• School Feeding in Zambia

• PMTCT in Rural Malawi

• Targeting Foodin WesternKenya

• Participatory Approach in Uganda

May 2005 Issue 25 ISSN 1743-5080 (web)

Special focus on Food Aid and HIV/AIDS

his special issue of Field Exchange isdedicated to Professor John Kevany (ENNDirector) who died just over two years ago(20th April 2003), as a tribute to his piviotalrole in establishing the ENN, his invaluablesupport until his untimely death, and hisinfluential work in the field of HIV/AIDS.

3

Contents

From the Editor

This special issue of Field Exchangefocuses on the food aid component ofHIV related programming and wasmade possible through additional fund-

ing from DFID RSA. How the ENN have gath-ered field material for this issue marks a signif-icant departure from our usual approach, inthat a consultant (Mary Corbett) was taken onas a kind of roving researcher/correspondent.Mary visited five countries over the course ofsix weeks (Malawi, Zambia, Kenya and Ugandaand South Africa) to meet with a wide varietyof agency staff implementing HIV programmingwith a food aid component. Her brief was todescribe these programmes and, where possi-ble, identify lessons learnt. She was also askedto identify significant related research initia-tives or findings in the countries visited. Anaccount of the experience is captured in hereditorial (page 17).

The rationale for this special issue was pri-marily that HIV-related programming involvingfood aid has been increasingly rolled out overthe past few years, with a view to achieving avariety of objectives. Since much of this pro-gramming is ‘cutting edge’, many of the objec-tives have not been properly tested. Henceagencies are, effectively, learning by doing. TheENN (along with others) believes it importantto document these new programming experi-ences to support a process of learning.Although the production of this special issue,based on 18 programme experiences, cannotbe described as a comprehensive overview, itis, at least, a snap-shot of what is going on.The ENN believe that some of the lessons andcross-cutting issues to emerge from this specialissue are important and indicate priorities andstrategies for the future.

The agency field experiences documentedare based on programmes implemented by awide spectrum of agency types, i.e. UN, INGO,local NGO and CBO. Some of the CBOs literallystarted out as spontaneous community initia-tives, led by a few dynamic individuals affecteddirectly or indirectly by HIV. Over time, theseorganisations have professionalised, expanded,and achieved a significant profile and statuswith international agencies who now collabo-rate with them, and often provide resourcesupport. The food aid and nutritional compo-

APPLE Aids Prevention and Positive Living Programme

ART Anti-retroviral therapyARV Anti-retroviralCCA Community Counselling Aide CBO Community based organisationCHBC Community Home Based CareCHS Community Household SurveillanceCHW Community Health WorkerCI Chronically IllCMC Community Management

CommitteesC-Safe Consortium for the Southern

Africa Food EmergencyCST Coping Strategy IndexCTC Community Therapeutic CareDOT Directly Observed TherapyEMOP EMOP Emergency OperationFFW Food for workHBC Home based careHEA Household Economy ApproachHH HouseholdHIV/AIDS Human Immunodeficency

Virus/Acquired Immune Deficency Syndrome

IGAs Income generating activitiesINGOs International NGOsIP Implementing partners

JEFAP Joint Emergency Food Aid Programme

MT Metric tonneNAC National Aids CommissionNGOs Non-governmental organisationsNNGOs National NGOsNRUs Nutrition Rehabilitation UnitsOI Opportunistic infectionsOTP Outpatient Therapeutic ProgrammeOVC Orphans and Vulnerable ChildrenPLWHA Persons living with HIV/AIDSPMTCT Prevention of Mother-to-Child

TransmissionPRRO Protracted Relief and Recovery

OperationRSA Rapid Situation AssessmentSC Stabilisation CentreSF Supplementary FeedingSFP Supplementary Feeding ProgrammeSLA Sustainable Livelihoods ApproachTA Tribal areaTA Traditional authorityVCT Voluntary Counselling and TestingUN United NationsVAC Vulnerability Assessment

CommitteeVDC Village Development CommitteeVAC Village Action CommitteeVRC Village Relief Committee

Acronyms

6 Research• WFP recipients’ weight gain at Reach Out

clinic• Impact of HIV/AIDS on household food

security and quality of life in Malawi• Integrating CTC and HIV/AIDS Support in

Malawi• HIV and child mortality• BMI: A Strong and Independent Predictor

of Survival• Rethinking food aid in the face of

HIV/AIDS• Early exclusive breastfeeding increases

HIV survival• Bangwe home based care project in

Malawi• Pregnant women’s uptake of antiretroviral

prophylaxis• Modification of complementary foods in

Zambia• Better practice in targeted food assistance• Cotrimoxazole as a prophylaxis for HIV

positive malnourished children• On-the ground perceptions of WFP food

assistance and PMTCT in Zambia

17 ENN in the Field• Editorial• Impact of HIV/Aids on acute malnutrition

in Malawi • Integrated PMTCT services in a rural setting

in Malawi• Integrated approach to supporting

chronically ill in Malawi• School Feeding programme in Zambia• REEP experiences in Western Kenya • REACH OUT food assistance in Uganda• Meeting Point: a local CBO in Uganda

31 News• UN draft guidance on programming• HIV/AIDS: A Guide for Nutritional Care and

Support • CRS conference in South Africa • Invite to join WABA HIV and Infant Feeding

Group• New e-resources on HIV and AIDS• Local resources for supporting PLWHA• Fact sheets on ARV treatment for NGO/CBO

staff• Community based technology to combat

HIV/AIDS

33 Views• South Africa HIV/Aids Pandemic• Support for PLWHA’S in Malawi

34 Letters

35 Field Articles• WFP HIV/AIDS programming in Malawi• Emmanuel International• HIV/AIDS and Food Security in Malawi• Nutritional support through HBC in Malawi• Three Ingredients of Success:Targeting Food

Assistance in Western Kenya• Participatory approach to food security in

Uganda• Evolution of GOAL activities in Malawi

44 Evaluation• Targeting vulnerable households within the

context of HIV/AIDS in Malawi • WFP Monitoring and Evaluation of HIV/AIDS

programming in Malawi

T

4

how food aid/nutritional support is effectivecannot be made. In some contexts wherethere is adequate infrastructure and staffcapacity, e.g. DOT and PMTCT, targets shouldbe established as a priority, so that data canbe rapidly generated to test the degree towhich objectives are being met. Also, byestablishing targets, it is more likely thatimplementing agencies will invest in collect-ing, collating and analysing data to demon-strate programme impact.

Findings from the ENN field work supportthe ubiquitous ‘impression’ that very littleimpact assessment is taking place. Most ofthe programmes visited had either not col-lected impact data, or had collected data butnot undertaken analysis. In many cases thiswas attributed to lack of time and resources,with some agencies admitting the need forprofessional support (e.g. Reach Out pro-gramme in Uganda). While this is a credibleexcuse for CBOs and local NGOs, it is lessunderstandable for INGOs or UN agencies.The absence of impact analysis is worryingand undoubtedly explains why there is cur-rently very little in the published literatureon this relatively new type of programming.The impact data that have been collected doappear to show some impact in terms ofweight gain1. However, much of these dataand analysis lack statistical rigor and areconceptually weak. This is not to be overlycritical. Impact assessment of this type ofprogramming is methodologically complex.For example, in a food aid/ARV or DOT pro-gramme, it would be necessary to isolate theimpact of food aid in terms of improving tol-erance to drugs, ensuring better and longercompliance with treatment, or reducingopportunistic infection by improving nutri-tional status. In the case of integrated foodsecurity programming, e.g. Oxfam and CRSprogrammes in Malawi, it would be neces-sary to control for a number of external fac-tors like rainfall/climate, agricultural inputsand initial wealth status, in order to drawconclusions regarding the role of food aid instrengthening longer term food security.Generally, there has been insufficient atten-tion given to impact assessment methodolo-gy for emergency related programming.Apart from the methodological challenges,there are also difficult ethical considerationsin impact assessment. Given the large num-bers of unknowns with respect to food aidand nutritional support in the context of HIVprogramming, there is an urgent need tostrengthen the ability of implementing agen-cies (especially CBOs and local NGOs) toundertake impact assessment.

Anecdotally, IP staff and beneficiariesindicate enormous value of food aid as acomponent of the many types of HIV relatedprogramming. Benefits cited touch on manyof the objectives claimed for these types ofprogramming, e.g. weight gain, improvedwell-being, less mortality, improved pro-gramme compliance, longer-term food secu-rity. With the exception of longer-term foodsecurity, there is no reason to disbelievethese findings, especially when one considersthat much of the targeting is to the poorestof the poor (see section on targeting).However, we have to do better than relyingon anecdotal reports of programme benefitsin order to strengthen understanding of howprogrammes are working (what is theprocess underpinning impact) so that pro-gramme design can be strengthened, and soas to increase belief in this type of program-ming - particularly in an environment wheredonors are increasingly sceptical of the bene-

fits of food aid/HIV programming.

Targeting criteria, stigma and resourcesA great deal of food aid targeting within

HIV programming currently takes place onthe basis of proxy indicators, i.e. chronic ill-ness, orphan containing households, singleparent/widow headed households, elderly-headed households, child headed households(child less than 18 years old) and householdswith children of deceased HBC clients. Thesecategories have been employed partly toaddress issues of stigma which is still amajor problem - particularly in Africa.However, there are important reservationsaround the use of proxy indicators. Forexample, what is the inclusion error if thepurpose is to target those with HIV? Do suchindicators really address issues of stigma(people aren’t stupid and quickly realise thatchronic illness probably means advancedstages of AIDS). Furthermore, and perhapsmore significantly, many argue thatHIV/AIDS cuts across all income classes sothat targeting the chronically ill or orphancontaining families is not an equitable meansof targeting resources. Although this issomewhat contradicted by the CRS Dedzaresearch in Malawi (see page 7), it is sup-ported by other studies, e.g. Seaman/Petty(Field Exchange 23) and Garnier/Situmu/Watkins (WFP /REEP, this issue). In a pilotprogramme in Malawi (Fisher and Munk),ACF employed a food requirement/dependen-cy ratio indicator, which the authors arguedis more equitable than proxy indicators likeCI or OVC. Another difficulty with the proxyindicator approach has been that, in the faceof limited food aid resources, many pro-grammes have had to make tough choices asthe number of CI or orphan containinghouseholds at village level has exceeded foodsupply, e.g. this has created tensions andconflict for village committees charged withtargeting responsibilities and implementingagencies. REEP in Kenya had to scale downfood aid programming during the hunger gapperiod. Some agencies, recognising theinequity of targeting on the basis of proxiesfor HIV - especially in extremely food inse-cure and chronically poor areas -have used atwo tier system, i.e. used proxies in conjunc-tion with indicators of economic/food securitystatus. Relying on the community is, clearly,key for economic or food security based tar-geting. Generally, in the programmes report-ed here, the community does target thepoorest of the poor.

Targeting food aid to PMTCT/ARV/DOTprogrammes is, undoubtedly, an efficientmeans of providing nutritional support to theHIV infected. However, although consensusregarding the differential nutritional require-ments and rations for those infected withHIV has not been reached, there is someagreement on the need for differentialrations according to stage of disease. Forexample, it is believed that a 10% increasein energy intake is required to maintainnutritional status and avoid weight loss ofasymptomatic individuals living with HIV,while those with AIDS related illnessesrequire a minimum of 20% increase in ener-gy intake. Further research is needed onoptimal protein requirements during thecourse of HIV disease, which may be

1 See field article, Nutritional support through HBC in Malawiand the research piece, WFP recipients’ weight gain at Reach Out clinic, in this issue.

2 See field article, REEP experiences in Western Kenya, by Mary Corbett

3 See field article, WFP HIV/AIDS programming in Malawi

nents of programming described here are inte-grated into several different programme types,i.e. DOT, HBC, CI/OVC, ARV, PMTCT and schoolfeeding. Most of these programmes have multi-ple objectives for the food aid component.Furthermore, although the programme write-upshave focused on the food aid element of progra-mming, many contain other sectoral elements aspart of an overall integrated programme package.For example, the CRS programme in Dedza,Malawi (p38) not only targeted food aid toPLWHA households, but also contained an IGAelement, as well as vocational training for olderorphans.

There are a number of key findings from thissnap-shot of agency programming:

Objectives and evidence of impactMultiple objectives for the food aid element

of programming are invoked. For example, the18 month food aid package as part of the PMTCTprogramme at St. Gabriels Hospital in Malawi(by Gertrude Kara and Mary Corbett) is meantto encourage compliance and ongoing educa-tional support, support abrupt weaning, ensurefull infant immunisation, ensure VCT for infantson reaching 18 months and provide an opportu-nity to target IGA. The community school feed-ing programme in Zambia (by Kate Vorley andMary Corbett) is aimed at improving enrolmentin HIV affected areas and dissemination of HIVrelated knowledge. The food aid component ofOxfam’s integrated programme in Malawi (p22)is intended to break the cycle leading to individ-ual and household crisis. While food aid may,indeed, be able to serve a myriad of objectiveswithin different programming contexts, it isnotable that almost all objectives (which eitherappear in programme documents or areassumed by programme implementers) areexpressed in general and qualitative terms, i.e.specific quantifiable targets are not set. Thus,where the objective is to increase weight gain orsurvival of HIV affected individuals, the actualpercentage weight gain or percentage increasein survival hoped for is never explicitly stated.As a result, achievement of objectives cannot betested. Indeed, in some cases and contexts, itmay be unrealistic to set specific targets, espe-cially where there is little prior programmeexperience. However, if targets remain nebulous,then progress in understanding whether and

Picture taken onWorld Aids Day

S.

Thurs

tans,

Mal

awi

Editorial

5

lines and conducting subsequent assess-ments) and then determining what propor-tion of households have achieved a targetlevel of food security following the interven-tion.

While discharge may be far easier for TBand PMTCT type programmes as these havea natural end point, e.g. completion of treat-ment or cessation of breast-feeding, adopt-ing exit criteria may remain problematic evenfor these, as many beneficiaries will effec-tively be discharged back to the poverty andfood insecurity from whence they came.Ideas of promoting food security in a 12 or18 month time frame are probably quiteunrealisable and certainly, as yet, untested.

There are also challenges about nutrition-al entry and exit criteria for this type of pro-gramming. Thus, in Malawi, MSF have beenfeeling their way with BMI cut-offs forARV/nutritional support and have adjustedthese on the basis of experience during theprogramme. However, MSF readily acknowl-edge that newly planned programmes for 7-14 year olds on ARV will raise fresh chal-lenges in terms of identifying appropriateanthropometric-based entry and exit criteria.We still lack experience of adult feeding pro-grammes targeted at HIV infected individu-als, especially with regard to length of time ittakes for improvement, proportion expectedto improve, optimal rate of improvement.This is even true for malnourished HIV posi-tive children. Work by ACF in Malawi showsthat not only were 30% of children admittedto NRUs HIV positive, but their outcome waspoor, with almost half discharged failing torespond.

Integrated programmingProvision of food aid on its own to PLWHA

is unlikely to have a marked impact on nutri-tion, morbidity, survival, food security, andgeneral quality of life. Although this is under-stood by most agencies, there are still alarge number of food aid and nutritionallyfocused HIV programmes where integratedprogramming is proving very difficult, (seeShoham. J, CRS Malawi, Oxfam Malawi, MSFLuxembourg) Research summarised in thisissue (CTC/VALID) shows the needs ofPLWHA and HIV affected households werefound to cover ten different categories.These categories were ranked by participantsunder two groups, primary (nutritional,material, medical, economic, psycho-social,spiritual and patient care) and secondary(food security, knowledge on prevention andvocational skills). The ranking of needsreflected a prioritisation of immediate overlong-term needs. More experienced and bet-ter resourced agencies like Oxfam have suc-cessfully established linkages with other pro-

grammes, e.g. IGA, but have still found ithard to strengthen linkages between HBCand health infrastructure. More medicallyfocused agencies like MSF, which workthrough health care infrastructure, havestruggled to integrate longer-term food secu-rity initiatives. What is clear is that fully inte-grated programming, whereby the multi-sec-toral needs of PLWHA are addressed in acoordinated fashion, is not something thatcan be established easily.

While scepticism still abounds regardingthe role of food aid in HIV programming,there is a need for well conceived and imple-mented pilot studies/programmes of multi-sectoral programming which can demon-strate how effective food aid can be withinan integrated package. These experiencesshould next allow for analysis of how toeffect integration and in which contexts goodintegrated programming can take place.Guidance material for implementing agenciesshould then follow. It may be that some formof mapping can subsequently take place atcountry level, to determine where integratedprogramming is feasible and to targetresources (including food aid) on this basis.The need to rationalise food aid and nutri-tional support programming to situationswhere there can be a significant and sustain-able impact (i.e. where a level of integratedprogramming can be assured) is key, giventhe logistical challenges posed by theupsurge of this type of programming in sub-sahelian Africa. The fact that this type ofprogramming is so decentralised i.e. imple-mented through small CBOs, schools, healthcentres, means that costs per tonnage offood delivered is relatively high compared tomore traditional types of food aid program-ming, e.g. general rations or supplementaryfeeding. Economies of scale no longer apply.Therefore, targeting food aid and nutritionalsupport resources to locations where there ismost likely to be an impact (where program-ming is integrated) will reduce inefficientlogistics. It is also likely to have a greaterinfluence in convincing donors that this typeof programming is worthwhile.

This special issue also carries a largenumber of research summaries (publishedand grey literature) on topics related to foodaid, nutrition and HIV programming.Subjects range from BMI as a predictor ofsurvival in HIV/AIDS affected adults, to therole of community based technology to com-bat the impact of HIV/AIDS on food securityand livelihoods. Other subjects include theimpact of food aid on survival and nutritionalstatus in Bangwe, Malawi and the prevalenceof HIV in children admitted to NRUs inMalawi comparing their recovery perform-ance to HIV negative children.

The range of research is a stark reminderof how much we don’t know, as well as howmany people/institutions are doing researchout there. A lot of this research may notmake the published literature. Indeed, therecent IFPRI conference in Durban (11th-13th of April)4 received an enormous num-ber of abstracts of ongoing research. Clearly,many of those who submitted theseabstracts will need support and funding toensure that correct research protocols arefollowed and that findings can be written upand disseminated.

ConclusionThere is an argument that ‘while there is

little empirical evidence regarding the effec-

increased due to nitrogen loss associatedwith opportunistic infections (for themoment, current guidelines (WHO, FANTA)advise to at least the meet protein intakerequired of a balanced diet). Agencies likeWFP need to monitor emerging consensusand expert opinion on this and adjust rationsaccordingly. At the same time, implementingration differentials will be extremely chal-lenging from a logistical perspective.

Finally, despite the concerns that target-ing PLWHA will cause problems of stigmatisa-tion, the evidence does seem to show thatCBOs and NGOs that invest in communitysensitisation manage to reduce stigma.Agencies like REEP appear to have done anextraordinary job in building self-esteem ofthose infected and enabling them to speakfreely about their status. The increase innumbers of those coming to be HIV tested inREEP programme areas is testimony to this.

Exit criteriaProvision of food aid within the context of

HIV programming brings with it the enor-mously challenging issue of when to stopgiving out food. Many of the programmesdescribed have not invoked exit criteria,(e.g. REEP2 and Shoham J3) and some bene-ficiaries have been food aid recipients foralmost three years, e.g. Reach Out. Althoughthere has been much discussion in the litera-ture, and within implementing agencies,about exit criteria and discharging recipientson achieving a level of food security, thereality is that there are no simple tools toassess when that has been achieved. Theonly food security assessment tools widelyemployed are HEA and, more recently, CSI.Although HEA appears a promising tool infood security impact assessment, theapproach requires extensive field basedtraining and cannot be rolled out rapidly andon a large scale, especially if the status ofindividual households is to be assessed.Furthermore, as many beneficiaries areextremely poor (indeed they are targeted onthis basis), the time-scale for achieving alevel of food security may be entirely unreal-istic for many of these programmes. In reali-ty, what is being proposed is a form ofdevelopment which, in many cases, is notsomething that can be squeezed into thetime-frame of a PRRO or the ‘attention span’of a donor food aid programme. Currently,and unsurprisingly, there are no data in thepublished or grey literature on proportion offood aid beneficiaries in this type of pro-gramme who are being discharged (exitingthe programme). Progress in this area canprobably only be made in an extremelypiece-meal fashion, with implementing agen-cies piloting food security impact assess-ments in a few locations (constructing base-

EditorialVan

essa

Vic

k/W

FP

6

4 See http://www.sahims.net/archive/specialfocus/specialcoverage_who_consultation2.htm

5 Kadiyala S and Gillespie S (2004). Rethinking food aid to fightAIDS. Food and Nutrition Bulletin, vol 25, no 3, pp 271-282.See research summary in this issue.

Any contributions, ideas or topics for futureissues of Field Exchange? Contact the editorialteam on email: office@ennonline.net

In October 2002, the medical records of 100clients who were registered to receive WorldFood Programme (WFP) food through theReach Out Clinic-Mbuya Parish HIV/AIDS

Initiative were reviewed. The programme wasbased in Kampala, Uganda.

The food disbursed was based on a family sizeof 5, and was intended to comprise 50% of the foodneeds of the family. It consisted of maize meal (25kg every 24 days), beans (25 kg every 45 days),corn-soya blend flour with micronutrient fortifica-

Both Antonia Torreblanca (MD, MPH) and EileenKim (MD) worked as physicians at the Reach Outclinic, Kampala, Uganda in November 2003. Theyare currently working as attending physicians atthe Kaiser Permente Medical Centre, California,USA.

This report outlines the weight progress of clientsattending the Reach Out clinic of Mbuya parish clin-ic in 2002, who were receiving WFP food (see fieldarticle p28).

WFP Recipients’Weight Gain atReach Out Clinic

tion (25 kg every 3 months), and cooking oil (4 Levery 2 months).

Table 1 outlines the baseline characteristics ofclients enrolled in the WFP-Reach Out pro-gramme. Of the 100 clients that began receivingfood in October 2002, 85 were alive at 1 year. Ofthose 85 clients, 35 (41%) had HIV, 49 (58%) hadHIV and TB, and one person (1%) had TB alone.

Table 2 reflects the trends in their weight priorto receiving food, and at 3 months, 6 months, and12 months after joining the programme. Duringthe four months prior to receiving WFP food,clients’ weights remained stable (average changeof +0.25kg, P=0.47). At 6 and 12 months, patientshad statistically significant weight gains (+2.3kg,P<0.001 and +2.4kg, P<0.001 respectively).Although all groups demonstrated weight gain,the most marked gains were in the clients withboth HIV and TB (see figure 1).

ConclusionsAlthough average weight gains were modest,

at 6 and 12 months nearly 50% of clients had sig-nificant weight gains of 3 or more kilograms (6.6lbs). Weight gain in the ‘HIV alone’ group peakedat 6 months and then began to decline. This maybe secondary to the progression of their HIV infec-tion. This study demonstrates a positive outcomefrom WFP food disbursements and proves the pro-vision of food to patients with HIV and/or tuber-culosis in impoverished nations should be sup-ported.

For further information, contact: AntoniaTorreblanca and Eileen Kim, The PermanteMedical Group, 280 West MacArthur Blvd,Oakland, CA 94611, USA. email: and Eileen Kim,email: antonia.torreblanca@kp.org andeileen.kim@kp.org

By Antonia Torreblanca and Eileen Kim

Table 1 Baseline characteristics of clients enrolled in the WFP-Reach Out programme

A. Characteristics of the 85 clients alive at 12 months

Characteristics HIV aloneN=36

TB aloneN=1

35 (+8.4)

Mean age, yr (SD)

35 (97%)

59 (+10.9)

8 (22%)

Female, no. (%)

57 (+13.2)

3 (23%)Started on HAART1

55(+0)

Table 2: Weight change according to disease status

DiseaseStatus

Wt change, 3 months

Wt change,6 months

Wt change,12 months

HIV+TB

HIV +0.07 kg (P=0.90)

+0.58 kg(P=0.11)

+2.06 kg(P<0.001)

OverallAverage

+0.25 kg(P=0.47)

+0.80 kg(P=0.011)

+2.33 kg(P<0.001)

+2.43 kg(P<0.001)

+ 0.40 kg (P=0.40)

+1.01 kg(P=0.044)

+2.49 kg(P<0.001)

+2.98 kg(P<0.001)

+1.68 kg(P=0.024)

Wt change,4 months to start

HIV aloneN=2

8 (62%)

Mean age, yr (SD)

HIV/TB N=13

Female, no. (%)

Mean wt, kg (SD)

Started on HAART1

Mean wt, kg (SD)

B. Characteristics of the 15 clients deceased at 12 months

Characteristics

36 (+6.7)

38 (79%)

54 (+9.7)

11 (23%)

36 (+9.6) 45 (N/A)

0 (0%)

0 (0%)

68 (N/A)

N/A

2 (100%)

37 (+2.1)

HIV/TB N=48

1 Highly Active Antiretroviral Therapy

Figure 1 Greater weight gain in clients with HIV and TB

62

60

58

56

54

52

500 3 6 9 12

Wei

ght

(kg)

Months

Researchtiveness of food aid in responding toHIV/AIDS, this should not forestall action’ andthat ‘a well documented learning-by-doingapproach is required of building up, evaluatingand disseminating experiences and lessonslearned’ (Gillespie S, 20045). However, theremay be dangers with such an approach. Whatthis issue of Field Exchange has shown is thatthere are poorly described objectives and lim-ited, if any, impact assessment of many pro-grammes using food aid within an HIV pro-gramming context. Indeed, impact assess-ment in the food security sector is notoriouslydifficult with tools and capacity poorly devel-oped. There are also huge challenges withregard to targeting PLWHA, not least of allwhether it is equitable to target foodresources on the basis of HIV status or someproxy indicator. Perhaps even more significantis the issue of when to stop food aid.Certainly, food aid is unlikely to do any signifi-cant harm (except maybe tie up implementingagency resources or undermine volunteerismwhere carers are targeted). However, by defi-nition, food aid can only be seen as a short-term resource transfer. Longer-term develop-ment objectives, which are increasingly beingclaimed for food aid, may in most situationsbe unrealistic, so that the exit criteria beinginvoked are unlikely to be applied. We need towatch this closely. There is something veryunconvincing about claims that two or threeyears of food aid, in conjunction with a wreathof livelihood initiatives, will help secure a sig-nificant shift in food security.

There are also urgent issues to beaddressed around poorly integrated program-ming, as well as the logistical challenges ofthis kind of decentralised food aid program-ming. Limited evidence (although it is proba-ble that more could be made available) sug-gest that logistical costs are high compared tomore standard emergency food aid program-ming. Furthermore, although we don’t havedata, it is likely that where programmes arepoorly integrated with other services, impactwill be limited. While there is little doubt thatmany very poor people are benefiting fromthe food aid – after all it is a resource trans-fer, and this explains why so many imple-menting agency staff and beneficiaries arepositive about the programmes – the principlequestion we should be asking is ‘could themoney involved be spent more effectively’? Atthis stage it would seem eminently sensible tostick with small–scale pilot programmes usingfood aid as part of an integrated package ofsupport, and to monitor and document theimpact food aid can have in different types ofprogramme and context. This may then placeus in a better position to target food aid effec-tively as part of HIV programming, ratherthan roll out large-scale programmes andhope to learn something from them after theevent. There is absolutely no reason to makefood aid the proverbial ‘cart’ before the horse.

Jeremy Shoham (editor)

asset sales and PLWHA presence on quality of lifeof the head of household. However, a significantmain effect on quality of life emerged within thevariable of asset sales.

LimitationsOne of the main limitations with most of the

research focusing on HIV/AIDS is the inability toappropriately and correctly identify people livingwith HIV or AIDS. Most research now uses ‘chroni-cally ill’ as a proxy indicator for HIV infection.Initial analyses from some researchers have indicat-ed that this proxy indicator over-exaggerates thecases of HIV infection. Barrère (2005)3 examined theuse of this proxy indicator in a survey in Malawiand found that only 54% of a chronically ill samplewas likely to have HIV or AIDS based on nationalinfection rates.

As such, the survey used in this study specifi-cally asked if anyone in the household was infectedwith HIV or AIDS. A total of 12.9% of the house-holds replied affirmatively. According to theNational AIDS Commission in Malawi (2003)4 , theinfection rate among rural households is estimatedto be 10-15%. Thus, the 12.9% of affirmativeresponses would fall as projected within the infec-tion rate estimates. However, it is possible thatsome households answered negatively to this ques-tion due to stigma concerns or lack of knowledge offamily members’ status, which could result in ahigher HIV positive population than the one pre-sented here.

Finally, this is a relatively small sample and can-not be applied broadly to the larger rural popula-tion in Malawi in other districts. These data wereobtained from a specific geographic location for thepurposes of an intervention evaluation, thus theresults are not necessarily applicable to a broaderarea.

DiscussionThe results demonstrate that the presence of

PLWHA and OVC (orphans and vulnerable chil-dren) at the household level also impact the level offood security, which is highlighted by whether thehouseholds have access to sufficient food to meettheir needs. This relationship is further compound-ed when households, lacking this supply, cope withtheir situation by selling their assets. These vari-ables, in turn, affect the overall quality of life of theheads of households.

The data from this evaluation highlight the needto engage in holistic programming. Traditional mit-igation responses, such as simple home-based careprogrammes, cannot respond to the overall stressplaced on the household in terms of food insecurity.Psycho-social interventions will also be hamperedwithout addressing the basic food needs, as qualityof life will likely decrease alongside assets and foodsecurity. In addition, given that these data demon-strate a link between the presence of OVC in thehousehold and the quality of life and food securityof the household, programming will need toaddress these findings in future OVC program-ming.

Given these results, programming will need tofully address mitigation of HIV/AIDS at the house-hold level. While many theories have aboundedregarding the impact of HIV/AIDS on householdfood security, this evaluation confirmed this link inrural, central Malawi, while uncovering the psy-chosocial impact of this relationship on the house-holds.

For further information, contact ShannonSenefeld email:ssenefeld@crssaro.org.zw or shan-nonsenefeld@yahoo.com

1 Dedza End of Project Evaluation, July 2003.2 Ferrans C, & Powers M (1992). Psychometric assessment of

the Quality of Life Index. Research in Nursing and Health, 15, 29-38.

3 Barrère, B. (2005). Pre-Test of New HIV Indicators. Presentation from ORC Macro, UNICEF, DHS & USAID. Washington, DC. January 6, 2005.

4 National AIDS Commission (2003). Estimating National HIV Prevalence in Malawi from Sentinel Surveillance Data: Technical Report: October 2003.

Recent research has suggested thatHIV/AIDS in southern Africa is directlylinked to increased household food insecu-rity. The research presented here is derived

from an evaluation study of Catholic Relief Services’(CRS) home-based care programme in one ruralarea of Central Malawi. The project commissionedan evaluation of interventions, including food secu-rity and outreach to households affected byHIV/AIDS, ending in September 2004.

Methods and designA total of 326 households participated in the

study. The households were selected using areaprobability sampling. Twenty villages were selectedin total. Of the households that participated, 25households were eliminated from the final analysesdue to incomplete responses or invalidated surveys.

Each head of household was administered a sur-vey in the local language. The survey was com-prised of three sections. The first was adapted fromthe Community Household Survey (CHS), whichhas been used widely in southern Africa to examinehousehold vulnerability. The CHS is a self-reportingmeasure that asks respondents to identify theirhousehold assets and food security levels. The corecomponents of the CHS remained intact for thisevaluation, and included questions on existing foodsupply and expected food security in times ofdrought. An additional section was added thatasked families knowledge-specific questions relatedto HIV/AIDS, in order to evaluate the effectivenessof the HIV/AIDS awareness campaigns within theareas.

The second section asked for demographicinformation, including the age of the head of house-hold and the number of children and orphans with-in each household. This section specifically request-ed information on HIV/AIDS within the surveyedhouseholds. The final section was a Quality of LifeIndex (QLI), which was selected due to its compre-hensive psychosocial nature. The QLI is a standard-ised measure, developed by Ferrans and Powers2.While this measure has been standardised amongvarious populations around the world, it has notbeen used before within Malawi. However, psycho-metrics revealed that among the sample population,the QLI performed similarly to past performancewith other populations.

ResultsA total of 301 household surveys were analysed

in the results. The sample represents 20 villages ran-domly selected from three deaneries. The averageage of the head of household was 45.7 years. Femalehead of households constituted 40.1% of the house-holds. Approximately 3.4 children were reported,on average, for each household, while an average5.6 people were reported as residing in the house-holds. Just under half (49%) of households reportedthat an orphan resided in their homes, but only23.2% of households reported the presence of AIDSorphans in their households.

Of the sample, only 12.9% of the householdsreported having someone who was living withHIV/AIDS within their households, but 45.5%reported benefiting from a home-based care pro-gramme. While home-based care programmes alsocover chronic illnesses other than HIV/AIDS, theprimary service delivery is around HIV/AIDS relat-ed illnesses. This suggests that the sample either didnot know their family members may have beeninfected, or chose not to reveal their family mem-ber’s status to interviewers. However, more than65% of the sample indicated that they knew wherethey could access voluntary counselling and testingservices, and more than 90% indicated that theywould welcome and care for a family member whowas infected. Further data analyses revealed a neg-ative relationship between households who report-ed the presence of someone infected withHIV/AIDS and the self-reported willingness to dis-close status variable (p<0.05).

More than one-third (37.7%) of householdsreported selling assets in the last three months. Ofthe 107 households that reported selling assets, theprimary reason for doing so was to meet householdfood needs (40.3%), followed by the need to meetdaily household expenses (29%), and the need tocover hospital and doctor bills (11.2%). On average,households reported that in a non-drought year,they would only be able to meet their householdfood needs for 4.5 months with their current harvest(SD=2.7).

There was a significant difference (p<0.05) inreported food security between households thatreported having a household member living withHIV/AIDS and non-affected households. A relation-ship (p<0.01) also emerged between the presence ofAIDS orphans within the household and a reducednumber of months per year the household couldmeet its food needs. In addition, the number ofmonths that households could meet their food sup-ply with their current harvest was correlated signif-icantly with the participation of the households inthe HBC project (p<0.05).

Analyses demonstrated quality of life was pre-dicted by whether or not the household had soldassets in the previous three months (p<0.001) andwhether their current harvest food supply wasabove or below average supplies (p<0.017).Decreased quality of life scores were also signifi-cantly associated with the presence of persons livingwith HIV/AIDS (PLWHA) (p<0.05) and the pres-ence of orphans (p<0.05) in the household.

Computing a two-way analysis of variance formain effects of asset sales and the presence of anHIV household member on quality of life, demon-strated that there was no interaction effect between

Impact ofImpact ofHIV/AIDS onHIV/AIDS onhousehold household food securityfood securityand quality and quality of life inof life inMalawiMalawi

By Shannon Senefeld,Catholic Relief Services

Summary of evaluation1

Shannon Senefeld is the RegionalTechnical Advisor, HIV/AIDS withCatholic Relief Services for theSouthern Africa Region

The author would like to acknowledge the contri-butions of Catholic Relief Services-Malawi andCADECOM Dedza-Malawi to the work reflected inthis article.

Research

Double-croppingmaize and beansas part of Dezdaagri programme

Sewing classesas part of

orphan skillstraining in CRS

programme

M Corbett, ENN, 2005

M C

orb

ett,

EN

N,

2005

7

8

Over the last five years, Valid Internationalhas spearheaded the development andimplementation of the Community-basedTherapeutic Care (CTC) approach for the

treatment of acute severe malnutrition. The CTCmodel is designed to provide timely, effective andcost-efficient assistance in a manner that strengthensand empowers the affected communities and cre-ates a platform for longer-term solutions to theproblems of food security and public health. Inpractical terms, the combination of its three corecomponents (Stabilisation Centre (SC), OutpatientTherapeutic Programme (OTP) and SupplementaryFeeding (SF)) with food security and health educa-tion programmes, offers a more holistic approach tothe treatment of malnutrition.

CTC was first implemented in Malawi as a jointpilot programme between Valid International andConcern Worldwide during the 2002 nutritionalemergency. The focus of the CTC programme inMalawi has, since the end of the emergency period,shifted to a more long-term integration with localhealth structures. The programme has proved suc-cessful, with over 3,000 severely malnourished chil-dren admitted since 2002, and a 69.4% recovery rateas of early 20042 . In spite of the encouraging results,and the support that the approach has gained with-in country, the impact of HIV/AIDS in Malawi hasforced a more in-depth re-examination of the apti-tude of the current CTC model to face the so-called‘New Variant Famine’ and the HIV-fuelled increasein food insecurity and malnutrition rates in Africa3.

Since January 2004, Valid International, in col-laboration with Support and Analysis for Researchin Africa (SARA) and Food And Nutrition TechnicalAssistance (FANTA), has been conducting a two-part study to explore ways in which existing CTCprogrammes can be adapted to provide community-based care and support to HIV-affected individuals,households and communities. The findings of thefirst, more qualitative part of the research are pre-sented in this article. They offer the first insight intothe opportunities, challenges and long-term impli-cations of such an essential expansion of the CTCapproach.

Aims and objectives of the researchThe research study is divided into two phases.

The first, phase A, explores the use and appropriate-ness of CTC as an entry point for providing longer-term HIV-related care, treatment and support in thecommunity. The second stage, phase B, aims toestablish whether HIV positive children can recoverfrom severe acute malnutrition when given a stan-dard CTC protocol. The findings outlined in thisarticle relate exclusively to phase A of the study.

The aims of phase A are three-fold:

a) to examine the effectiveness of the CTC programme in Dowa district for identifying households affected by HIV/AIDS

b) to examine the relationship between the perceived needs of, and what is being provided for, People Living With HIV/AIDS (PLWHA) and HIV affected households in Malawi, and

c) to examine how well the CTC programme in Dowa district meets both the perceived and actual needs of PLWHA and HIV affected households.

The research focused primarily on Dowa dis-trict, but activities were also conducted in Lilongwe,Blantyre and Nkhotakota districts.

Given the qualitative nature of this part of thestudy, the methodology relied heavily on FocusGroup Discussions (FGDs), Semi-StructuredInterviews (SSIs), Questionnaires and Surveys.Participants included support agencies (e.g. non-governmental organisations (NGOs), Community-

Based Organisations (CBOs), governmental groups,community support groups, etc.) beneficiaries of HIVsupport programmes, members of HIV affectedhouseholds, traditional health practitioners, agricul-tural extension workers, Ministry of Health (MoH)staff and traditional local leaders, among others.Planning, conducting and analysing the results ofthis first phase of the study extended over a six-month period.

General findings The majority of the support programmes for HIV

affected households surveyed during this study werefound to rely on proxy indicators for identifying ben-eficiary households. Of these indicators, the mostcommonly used are chronic illness (i.e. 3-12 months),households looking after orphans (both parentsdeceased), single parent/widow headed households,elderly-headed households, child headed house-holds (child less than 18 years old) and householdswith children of deceased HBC clients.

The needs of PLWHA and HIV affected house-holds were found to cover ten different categories.These categories were ranked by participants undertwo groups, primary (nutritional, material, medical,economic, psycho-social, spiritual, and patient care)and secondary (food security, knowledge on preven-tion, and vocational skills). The ranking of needsreflected a prioritisation of immediate over long-termneeds, as well as the gaps in the support available inthe research area.

The support available to PLWHA and HIV affect-ed households was found to come from both formal(e.g. international NGOs, CBOs, governmental struc-tures) and informal sources of support (e.g. commu-nity initiatives and groups). Formal sources weremost effective at providing nutritional, medical, foodsecurity and economic assistance. Informal sources,on the other hand, were most successful at address-ing the spiritual, psycho-social and patient carerequirements of PLWHAs and HIV affected house-holds. Together, both informal and formal sourcesengage with all but one of the primary areas of needand one of the three secondary needs. Evidence col-lected during the study, however, suggests that theprovision of support by formal sources – such asNGOs – has led to the weakening of informal sup-port mechanisms within the communities. InMalawi, the support of community members is oftencrucial in dealing with illnesses in the household.One of the study participants said: “I have been sickfor three years and the assistance I have received hasbeen enough, given that the people who assist me arethemselves poor but spare something for me…theywash my clothes, beddings and give me other thingsthat I need in my household”. The availability ofexternal formal support, however, is weakening thistype of community assistance. In the words of anoth-er respondent, “some people in the community saythat we get a lot of money from [the NGO] so theyrefuse to assist us”. The effect of formal support pro-grammes has also been felt in NGO initiatives tostrengthen food security in the communities at large.The study found evidence that the singling out ofindividual households for HIV related support, leadsto dissatisfaction and decreasing levels of communi-ty participation in other programmes aimed at thecommunity at large.

Although formal and informal sources of supportjointly take on most need areas of PLWHA and HIV-affected households, the study concluded that theappropriateness of the assistance available to thesehouseholds is limited by inadequate engagementbetween support providers - to reduce replication -and between providers and beneficiaries. Initialengagement - in the form of needs assessments, forexample - was inadequate, resulting in a poor under-

IntegratingCTC andHIV/AIDSSupport inMalawiBy Saul Guerrero, Paluku Bahwere, Kate Sadler, and Steve Collins, Valid International

Saul Guerrero is a Social and CommunityDevelopment Advisor.Dr. Paluku Bahwere (PhD) is a medical doctor. Kate Sadler is a Public Health Nutritionist.Dr. Steve Collins (MD) is a medical doctorand Director of Valid International.

This article describes the research findings1 of thefirst phase of a two-part study in Malawi by ValidInternational, which is exploring how existing CTCprogrammes can be adapted in the context ofHIV/AIDS.

The authors regularly work as CTC pro-gramme advisors with Valid International,and were involved in the coordination andconduction of the CTC and HIV/AIDSstudy in different capacities.The authors would like to thank Food andNutrition Technical Assistance (FANTA)and the SARA project for financial andtechnical support, and Concern Worldwideand Ministry of Health and Population(MoHP) in Dowa District for facilitatingthe conduction of the study.

1 Valid International/SARA/FANTA, 2004. Study to Examine the use of Community-based Therapeutic Care (CTC) to support HIV/AIDS infected and affected individuals, households and communities. Valid International, Oxford, 2004

2 Collins, Steve, 2004. Community-based Therapeutic Care; A New Paradigm for Selective Feeding in Nutritional Crises. HPN Network Paper, No. 48, London, November 2004

3 De Waal, Alex & Whiteside, Alan, 2003. New Variant Famine: AIDS and Food Crisis in Southern Africa, The Lancet, 362: 1234-37, October 2003

Research

Locally produced RUTF isprovided to beneficiaries

in the CTC programme

Val

id,

Mal

awi

9

standing of the most basic needs of the individualsand households targeted (e.g. beneficiaries receiv-ing tools and seeds when their priority needs relat-ed to hygiene and health). Post-distribution evalua-tions and the erratic distribution of goods and sup-port also resulted in limited health impact andwidespread dissatisfaction among beneficiaries.

Stigma was found to be a cross-cutting issuethroughout the conduction of this study. Althoughthere are indications that stigma may be on thedecrease at a national level – as a result of awarenesscampaigns and the better availability of support – inDowa district, stigma continues to surroundHIV/AIDS, affecting the identification and provi-sion of support to those infected and affected by thedisease. Although stigma surrounding HIV/AIDShas been traditionally associated with a lack ofunderstanding of the nature and transmission of thedisease, the study also found links between stigmaand the provision of support by externalagencies/organisations. In the words of one respon-dent, “…because we came out in the open as HIVpositive, people say that we should not be givenassistance because we are going to die soon.”

The current CTC model and HIV/AIDS inMalawi

Using the most commonly used proxy indica-tors described above, the study established that arelatively high proportion (65.7%) of HIV affectedhouseholds with malnourished children hadenrolled in the CTC programme. Due to the largenumber of HIV affected households without mal-nourished children, only 16.2% of the total HIVaffected households surveyed were enrolled in theCTC. This indicates that in order to target a greaterproportion of the HIV affected population, the pres-ent CTC model would need to expand its target cri-teria and include other indicators beyond child mal-nutrition.

Under the current CTC model, all beneficiariesreceive nutritional, medical, food security and mate-rial support. The study concluded that – from acommunity perspective – the current CTC approachadequately meets the support needs of targeted chil-dren in these areas. The nutritional, medical, foodsecurity and material supports were deemed suffi-cient by participants, and the mode of delivery andquality of the distribution process widely charac-terised as appropriate. The participants’ viewsregarding the appropriateness of the medical andnutritional care are substantiated by the high recov-ery and low non-response rates in the programme.In terms of the adequacy of the support for thehousehold, however, the majority of the respon-dents felt it was insufficient to meet their house-hold’s needs. This was a foreseeable conclusion,given that the current targeting criteria focuses onindividual malnourished children and not on meet-ing the nutritional (or any other) requirement ofHIV affected households as a whole.

Future implications for the CTC approachAlthough the results from the initial phase of

the study shed some light on the changes requiredfor the CTC to be used as an approach for thedelivery of HIV support, the overall findings of theresearch – i.e. from phases A and B – will be neces-sary before conclusive steps are taken towards anyadaptation of the CTC model. So far, however, thestudy has shown that in order to target a muchgreater proportion of HIV affected households,changes in the current targeting criteria of the CTCprogramme are required. CTC must expand its tar-get group to include malnourished and chronical-ly ill adults. Current anthropometric indicators foridentifying malnutrition in adults and childrenmust also be used alongside a combination ofsome of the proxy indicators identified in thisstudy (and further explored in phase B). Thesechanges would ultimately allow for more effectiveidentification and the inclusion of a larger propor-tion of HIV affected households.

The findings also suggest that the CTC as anapproach is suitable for providing longer-term careto HIV affected and infected individuals andhouseholds. CTC was found to be advantageousdue to its minimal impact on the daily activitiesand resources of the enrolled households, anddecreased risk of contracting opportunistic infec-tions. The assistance currently provided under theCTC was reported as adequate to meet the needsof targeted children residing in HIV affectedhouseholds. The medical and nutritional protocolappear also to be effective at treating malnour-ished HIV-infected children, an area which is cur-rently being further assessed under phase B. Somechanges and additions, however, are necessary to

HIV contributes substantially to childmortality, but factors underlyingthese deaths are inadequatelydescribed. With individual data from

seven randomised mother-to-child transmission(MTCT) intervention trials, a research team esti-mated mortality in African children born to HIV-infected mothers and analysed selected risk fac-tors.

HIV and ChildMortalitySummary of published research1

Early HIV infection was defined as a positiveHIV-PCR test before 4 weeks of age and late infec-tion by a negative PCR test at or after 4 weeks ofage, followed by a positive test. Mortality ratewas expressed per 1000 child-years. The teaminvestigated the effect on mortality of maternalhealth, infant HIV infection, feeding practices,and age at acquisition of infection on mortality.This was assessed using Cox proportional haz-ards models, and allowed for random effects fortrials grouped geographicaly.

Overall, 378 (11%) of 3468 children died. Byone year of age, an estimated 35.2% infected and4.9% of uninfected children will have died and by2 years of age, 52.5% and 7.6% will have died,respectively. Mortality varied by geographicalregion and was associated with maternal death(adjusted odds ration 2.27, 95% CI 1.62-3.19),CD4+ cell counts <200 per micro litre (1.91, 1.39-

enable the CTC to be a more effective medium for thedelivery of support to PLWHA and HIV affectedhouseholds. For example, the CTC needs to providelong-term HIV specific nutritional support to the HIVinfected, HIV-related nutritional counselling, andmore comprehensive palliative care.

Successfully using the CTC to provide support toPLWHA and HIV affected households, however, restsas much on the expansion of its own services as itdoes on the successful integration of existing sourcesof (formal and informal) support to form a coordinat-ed network of assistance. The information collectedduring this study offers initial suggestions for the cre-ation of such a network in a way that (re)empowersthe community, minimises replication and maximisesthe quality of the assistance provided to PLWHA andHIV affected households. The role of the CTC in thismultilateral approach could eventually be two-fold,as the developer of a comprehensive support packagethat incorporates the needs identified during thisstudy and WHO/UNAIDS recommendations, andby acting as a coordinating structure working toensure that support provided by partners is regularand consistent. This reconceptualisation of the CTC’srole is in its early stages. Over the next two years pre-cisely this model will be tested in a joint programmebetween Valid, the Zambian Ministry of Health andConcern Worldwide, designed to merge key aspectsof the CTC and the provision of support to HIVinfected and affected individuals and households.

For further information, contact Saul Guerrero,email: saul@validinternational.org or the ValidInternational office, email: office@validinterna-tional.org

2.62) and infant HIV infection (8.16, 6.43-10.33).Mortality was not associated with either everbreastfeeding and never breastfeeding in eitherinfected or uninfected children. In infected chil-dren, mortality was significantly lower for thosewith late infection than those with early infection(0.52, 0.39-0.70). This effect was also seen inanalyses of survival from the age at infection(0.74, 0.55-0.99).

The authors of the study concluded that thesefindings highlight the necessity for timely anti-retroviral care for HIV-infected women and chil-dren in developing countries, and for prophylac-tic programmes to prevent MTCT.

1 Newell ML et al (2004). Mortality of infected and uninfect-ed infants born to HIV-infected mothers in Africa: a pooled analysis. The Lancet, vol 364, Oct 2nd 2004, pp1236-1243

Research

Valid and MoH workers inDowa District

Valid, Malawi

10

The promotion of exclusive breastfeeding (EBF) to reduce the post-natal transmission (PNT) of HIV is based on limited data. Therecently published ZVITAMBO study in Zimbabwe has carefullydefined, measured and categorised infant feeding patterns, linking

this to HIV status and health outcomes. Non-breastfed infants were notincluded in the study. Of note, while HIV tests were done on these women,most chose not to obtain the results – which was the norm when the studywas conducted a few years ago. Thus this study is effectively measuringbehaviour among HIV+ mothers who do not know their status, which iscurrently the situation for the vast majority of mothers who are HIV+2 .

In the context of a trial of postpartum vitamin A supplementation, edu-cation and counselling about infant feeding and HIV was provided whileinformation was prospectively collected on infant feeding practices, andmeasured associated infant infections and deaths.

A total of 14 110 mother–newborn pairs were enrolled, randomlyassigned to vitamin A treatment group after delivery, and followed for 2years. At baseline, 6 weeks and 3 months, mothers were asked whetherthey were still breastfeeding, and whether any of 22 liquids or foods hadbeen given to the infant. Breastfed infants were classified as exclusive, pre-dominant, or mixed breastfed.

A total of 4495 mothers tested HIV positive at baseline. Of these, 2060of their babies were alive, polymerase chain reaction negative at 6 weeks,and provided complete feeding information. All infants initiated breast-feeding. Overall PNT (defined by a positive HIV test after the 6-week neg-ative test) was 12.1%, 68.2% of which occurred after 6 months.

Compared with EBF, early mixed breastfeeding was associated with a4.03 (95%CI 0.98,16.61), 3.79 (95%CI 1.40–10.29), and 2.60 (95% CI1.21–5.55) greater risk of PNT at 6, 12, and 18 months, respectively.Predominant breastfeeding was associated with a 2.63 (95% CI 0.59–11.67),2.69 (95% CI 0.95–7.63) and 1.61 (95% CI 0.72–3.64) trend towards greaterPNT risk at 6, 12, and 18 months, compared with EBF.

The authors conclude that EBF may substantially reduce breastfeeding-associated HIV transmission.

Recently produced training materials (Module 2 for health and nutrition workersin emergency situations) detail support of exclusive breastfeeding in the challeng-ing environment of emergencies. They support Module 1 material, both of whichare available in print from ENN, and online at http://www.ennonline.net

1 Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. Peter J. Iliff, Ellen G. Piwoz, Naume V. Tavengwa, Clare D. Zunguza, EdmoreT. Marinda, Kusum J. Nathoo, Lawrence H. Moulton, Brian J. Ward, the ZVITAMBO study groupand Jean H. Humphrey AIDS 2005, 19, 69-708. The full text version of this article is available by joining the WABA HIV and Infant Feeding Yahoo Group (see news piece this issue)

2 Comments drawn from Pro-Nut HIV communication, Ted Greiner, 14 April 2005

Early ExclusiveBreastfeeding IncreasesHIV SurvivalSummary of published paper1

Arecent detailed review of the relevant literature and thefindings of a mission to eastern and southern Africahighlights the implications of the HIV/AIDS pandemicfor food aid strategy and programming. The review

examines experiences of using food aid within the context ofPrevention of Mother to Child transmission (PMTCT) pro-grammes, food supplementation in home based care, supplemen-tary feeding for orphans and vulnerable children, food for educa-tion and training, food for work, and income generating activitiesand micro-credit schemes. A number of conclusions are drawn.

HIV/AIDS demands a multi-pronged response, grounded inan understanding of the susceptibility and vulnerability of peo-ple’s livelihoods. Food aid can strengthen care, mitigation andprevention activities. It is also relevant for the nutritional wellbeing of vulnerable groups and for strengthening human capital,as well as for preserving assets and livelihoods. Food aid canenable the marginalised to take advantage of development oppor-tunities. Those affected by HIV/AIDS are arguably among themost marginalised populations, both socially and economically.Not only does stigma foster exclusion, but progressive assetdepletion may also render households destitute and unable toparticipate in the development process. There are opportunitiesfor using food aid to enable these populations to avoid and escapesuch marginalisation. In addition, it will be important to seekopportunities for linking such initiatives to HIV-specific interven-tions wherever possible, thus potentially further reducing suscep-tibility to HIV.

But many challenges exist which will need to be dealt withdynamically through ongoing implementation, good monitoring,and timely focused operational research. Such challenges includehow to target the vulnerable in the HIV/AIDS context, how to usefood aid to leverage longer-term livelihood options, how toensure complementary resources through appropriate partner-ships and how to strengthen local capacity. Donor responses havebeen piecemeal to date, and the involvement of food aid organisa-tions is fairly recent. The article concludes that, “there is littleempirical evidence regarding the effectiveness of food aid inresponding to HIV/AIDS currently, this should not forestallaction. A well documented learning-by-doing approach isrequired, of building up, evaluating and disseminating experi-ences and lessons learned”.

1 Kadiyala S and Gillespie S (2004). Rethinking food aid to fight AIDS. Food andNutrition Bulletin, vol 25, no 3, pp 271-282

Rethinking FoodAid in the Face ofHIV/AIDSSummary of published paper1

The identification of basic prognostic indi-cators of HIV infection is essential beforewidespread antiretroviral therapy can beimplemented in low-technology settings.

A recent study assessed how well body massindex (BMI:kg/m2) predicts survival.

BMI within 3 months of HIV diagnosis wasobtained from 1657 patients aged >15 years,recruited in a seroprevalent clinical cohort in TheGambia since 1992, and followed up at least once.Baseline CD4+ counts and clinical assessment attime of diagnosis were collated.

The mortality hazard ratio (HR) of those witha baseline BMI <18 compared with those with abaseline BMI > 18 was 3.4 (95% CI, 3.0-3.9). Themedian survival time of those presenting with aBMI <16 was 0.8 years, in contrast to a mediansurvival of 8.9 years for those with a baselineBMI >22. Baseline BMI <18 remained a highlysignificant independent predictor of mortality

BMI:A Strong andIndependentPredictor ofSurvivalSummary of publishedresearch1

after adjustment for age, sex, cotrimoxazole pro-phylaxis, tuberculosis, reported wasting at diag-nosis, and baseline CD4+ cell count (adjusted HR= 2.5, 95% CI 2.0-3.0). Sensitivity and specificityof baseline BMI <18 was comparable to that of aCD4+ count <200 in predicting mortality within 6months of diagnosis.

BMI at diagnosis is a strong, independentpredictor of survival in HIV-infected patients inWest Africa. In the absence of sophisticated clini-cal and laboratory support, BMI may also provea useful guide for deciding when to initiate anti-retroviral therapy.

1 Marianne AB van der Sande, Maarten F.Schim van der Loeff, Akum A. Aveika, Saihou Sabally, Toyin Togun, Ramu, Sarge-Njie, Abraham S. Alabi, Assan Jaye, Tumani Corrah, and Hilton C. Whittle, (2004). Body Mass Index at Time of HIV Diagnosis. A Strong and Independent Predictor of Survival. J Acquir Immune Defic Syndr 2004;37:1288-1294

Research

11

The Bangwe project is a joint home basedcare (HBC) project run by the SalvationArmy and the Department of CommunityHealth, College of Medicine, University of

Malawi. While providing a standard HBC service toa township of 40,000 adjacent to the city of Blantyre,data have been collected since January 2003 onhealth problems of patients, their response to treat-ment and their nutritional status. Antiretroviral(ARV) drugs have not been given during the studyperiod.

Following a recommendation from the NationalAIDS Commission (NAC) that the project shouldassess the use of supplementary feeding to HBCpatients, WFP began providing food in July 2003.Nutritional assessments were carried out at the timeof initial assessment of the patient, and in June 2003,November 2003 and July 2004 on the members ofthe household of each patient.

Inclusion criteriaInclusion criteria were adult patients (over 15

years of age) with chronic disease of more than onemonth and in need of home based care. The groupof patients who were enrolled between January andJune 2003, and their families, did not receive food inthe first period of their home based care. They, andall subsequent patients, received the basic foodpackage from July 2003 to July 2004. The effect of thefood was measured by the difference in nutritionalstatus of each patient, comparing the Body MassIndex (BMI)2 at the latest survey assessment withthe original BMI, expressed as the rate of change per100 days.

The WFP supplement was targeted at house-holds taking care of orphans and those with some-one requiring HBC. Monthly rations were 50 kg ofmaize, 5 kg of beans and 7.5 kg of Likuni Phala (aMalawian fortified blended food). Four litres of oilwere given to half of the households selected on arandom basis.

The study was divided into three periods(January 2003 to July 2003, August 2003 to

November 2003 when the second nutritional surveywas implemented, and mid November 2003 to July2004, when the third nutritional survey occurred).Results were analysed using SPSS.

ResultsBetween January 2003 and July 2004, 360

patients were enrolled, of whom 59% were women.It appears that half of the chronically sick in thestudy area were enrolled at the time. The mean ageof men was 33.4 years and of women 30.9 years.Over the course of the study, over half of thepatients (56%, 199/360) died. There was no appar-ent difference in case severity of patients enrolled inthe different periods of the study based on symp-toms of fever, cough, lower limb pain and thrush.The majority of patients presented in an advancedstage of disease, with 70% in stage 4.

Change in BMIThe mean BMI at presentation was 18.5. Half of

the patients were malnourished, with a BMI lessthan 18.5 on enrolment, and one quarter wereseverely malnourished (BMI=16). The nutritionalstatus of the group presenting before July 2003 wassimilar to that of the group who presented duringthe second and third periods. During the first peri-od, when none of those enrolled received food,nutritional status remained constant. By the secondsurvey, the mean BMI of those still alive rose by 0.49per 100 days, and by 0.46 per 100 days by the timeof the third survey. These increases were not statisti-cally significant. For the small group of patients(n=22) who survived through all surveys, there wasan increase in the rate of change of BMI between thepre-food and the first post food period, but not withthe third study period. The addition of oil to thefood package had no effect on nutritional status, asmeasured by change in BMI per 100 days.

SurvivalOne third of patients died within four months of

being first seen, while half of those enrolled sur-vived 14 months. There was no difference in the sur-vival patterns of those who did not initially receivefood compared to those who received food from thestart. Survival was better in those allocated toreceive oil and those who actually received oil com-pared to those who did not. Oil seems to have aneffect, but only for those who survived six monthsfrom the time of initial assessment. The results showno statistically significant difference between thegroups before and after food distribution, althoughthere is a suggestion of improved survival in clinicalstage 4 patients post food distribution.

Household nutritionHouseholds of patients enrolled between

January and June 2003 (and measured in July 2003)were compared to households of patients measuredin late July 2004. Some of the families of patientsmeasured in July 2003 who survived a year werealso measured in 2004, and so were included in both

groups. Mean BMI fell between the two measure-ment dates, despite food supplementation from July2003. This pattern of a lower mean BMI of house-hold members persists when those surveyed in 2003are excluded from the 2004 group, and for differentage groups.

Discussion and conclusionsAn observational study of this sort is difficult to

interpret and there is much room for bias if similargroups are not compared. Using discriminant analy-sis of the presence and severity of presenting symp-toms, the ‘before’ and ‘after’ food groups had simi-lar mixes of case severity and comparable BMIs. Themain difference between the groups was the pre-ponderance of females in the ‘before’ food group. Itmay be that males tend not to seek HBC until it isknown that food is available. However, the severityof disease of those presenting when food was beingdistributed does not seem to differ from those pre-senting before the food handouts started. Overall, itappears that the two groups, at first presentation,are comparable.

In this study, food supplementation seems tohave no effect on survival, but does affect the nutri-tional status of those home based care patients whosurvived to one of the follow up weighing surveys.The result is not surprising, considering the latestage of disease in many presenting patients.Another possible reason for the absence of effect onsurvival could be that little food reaches the termi-nally ill, due to problems of distribution in an urbanarea of Malawi - families may have no one to carryfood home from the distribution point and manyneighbours are hungry. Oil, it was observed, mayhave an effect on those patients who survive sixmonths. This may be because it is a concentratedform of energy or because it is a saleable commodi-ty and money realised may be used to purchaseessential commodities.

The food supplement did not help maintain theBMI in household members of HBC patients. Thereduction in BMI of household members may beattributable to the socio-economic catastrophe ofloss of income and increase in expenditure, due tochronic ill health of one or two adults in the family.The longer the adult remains alive and ill, the longerthe loss of earnings, drain on resources and ensuingpoverty. This may account for the reduction in BMIin households of people living with HIV/AIDS,some of whose patients have survived for 12months or more. Arguably, the situation may havebeen worse without the provision of food.

1 An assessment of food supplementation to chronically sick patients receiving home based care in Bangwe, Malawi: a descriptive study. C Bowie, L Alinafe, R Marsh, H Misiri, P Cleary, C Bowie. Nutrition Journal 2005, 4:12 doi:10.1186/1475-2891-4-12 http://www.nutritionj.com/content/4/1/12

2 BMI calculated as weight (kg) divided by height in metres squared

Arecent study in Abidjan, Cote d’Ivoire setout to determine why pregnant women,who receive HIV-1 positive test resultsand are offered short courses of antiretro-

viral prophylaxis to prevent transmission of HIVfrom mother to child (PMTCT), do not participate innecessary follow up visits before starting prophy-laxis. The study was set up based on findings fromPMTCT work in Abidjan which found that during

Pregnantwomen’s uptakeof antiretroviralprophylaxisSummary of published research1

the programme’s first 15 months of operations(February 1998 to end of May 1999), 72% of 9657pregnant women who were offered HIV testingaccepted. Of the 884 women who tested positive forHIV-1, 395 received their test results. Only 118 (35%)of the 333 women who tested as positive for HIV,who received their test results and who were invit-ed to return for follow up visits during this period,eventually started taking zidovudine. Of the 215women who did not start taking zidovudine, 181had refused to return or discontinued follow up vis-its. Another 34 were lost to follow up or removedfrom the programme because they were ineligible.

In this study, a purposive sample of 27 womenwas selected. These women had received HIV-1 pos-itive tests and were invited to return for monthlyfollow up visits before starting prophylaxis withzidovudine at 36 weeks gestation, but had eitherrefused or discontinued the visits. None of thewomen started prophylaxis.

Most of the women explained their non-partici-pation in follow up visits by referring to negative

Bangwe HomeBased Care Projectin Malawi

Summary of published research1

experiences that they had had while interactingwithprogramme staff or to their views about the pro-gramme. Additional reasons concerned their disbe-lief of HIV positive test results and personal factors.

The study concluded that training and supervi-sion of programme staff may increase the likelihoodof positive interactions between staff and clients,thereby facilitating women’s participation in pre-venting transmission of HIV from mother to child.Outreach and mobilisation in communities that areserved by prevention programmes may comple-ment these measures at programme level, by con-tributing to increased social support for women’sefforts to prevent transmission of HIV from motherto child. Involving women’s partners in the preven-tion of mother to child transmission of HIV is wide-ly recognised to be desirable, but this rarely occurs.

1 Painter T et al (2004). Women’s reasons for not participatingin follow up visits before starting short course antiretroviral pro-phylaxis for prevention of mother to child transmission of HIV:qualitative interview study. BMJ volume 329, 4th September,2004, pp 343-345

Research

WFP warehouse

in Malawi

GO

AL,

Mal

awi

12

Poor quality complementary foods, charac-terised by high viscosity and low energydensity, and inappropriate feeding practiceshave been highlighted recently by the

World Health Organisation (WHO) as part of themajor causes of malnutrition in children. This isespecially true in the second semester of the firstyear of life in poor settings. Some of the approachesto improve the quality of complementary foodsinclude fortification and the enhancement of energydensity by the application of starch-breakingenzymes, such α-amylase.

A two-phased study has been undertaken inLusaka, Zambia since 2003, to assess whether mod-ification of complementary foods with α-amylaseand multi-micronutrient fortification benefit infantgrowth and micronutrient status. The study alsoaimed to generate data on breast milk intake of ninemonth old infants, using a technique where a stableisotope dose is administered to the mother. This wasa collaborative study that brought together theCentre for International Child Health (CICH),London, the Lusaka District Health ManagementTeam, the National Institute for Scientific andIndustrial Research (Lusaka) and Quality FeedsLimited (Lusaka).

MethodsThe study was based at Chilenje clinic, Lusaka,

a middle-income urban area where most house-holds have running water and flush toilets. Ethicalapproval was obtained from the University ofZambia Ethics Committee and Great Ormond StreetHospital, UK.

The first phase of the study, carried out in2003/2004, comprised an assessment of comple-mentary feeding practices among mothers of chil-dren aged 6-23 months old in Chilenje, Lusaka, andthe development and assessment of the acceptabili-ty of an α-amylase-modified complementary blendmade from locally produced cereals and legumes.

Complementary feeding practices were assessedby qualitative techniques using focus group discus-sions and formal interviews. Infants’ nutrient intakewas assessed by 24h–recall and 12h weighed foodrecords. Focus groups discussions were held withmothers, fathers and health workers at Chilenje clin-ic. A total of 34 mothers were interviewed, of whom

20 were observed at home to determine how theyprepared food and fed their children. Data from 24hand 12h weighed food records were used to com-pute the amounts of energy, iron and calciumobtained from family meals by children aged 6 -23months old.

The acceptability of the developed porridge wasassessed using sensory evaluation by 18 mothers atChilenje clinic. A blend was developed using maize,beans, groundnuts and bambaranuts and was treat-ed with α-amylase after roasting and hammermilling. Porridge viscosity was measured at differ-ent slurry concentrations (9-20%) in order to deter-mine the amount of additional dry flour possiblewithout a notable change in porridge consistency.The viscosity of habitual porridges was determinedafter simulating mother-reported and observedrecipes.

Acceptability of new blendThe results showed that maize (Zea maiz) is the

main cereal used in Lusaka. The available, andwidely used, legumes are common beans(Phaseolus vulgaris), groundnuts (Arachishypogea) and bambaranut (Vorandzea subter-ranean). Although commercially processed comple-mentary foods are available in Lusaka, they are veryexpensive, with the lowest priced selling at US$4per kg. Thus, lack of affordable complementaryfoods was found to be a major constraint on moth-ers’ feeding practices. On the contrary, the devel-oped blend would cost only US$2.5 per kg.Treatment of porridge with α-amylase allowed for a100% increment in porridge slurry concentrationwithout a change in porridge viscosity. The studyblend was widely accepted by mothers.

Micronutrient status and growthIt was observed that although children received

the recommended amounts of energy from habitualfoods, they received less than half of their dailyrequirements for iron and calcium. This highlightedthe need to fortify complementary foods to meet themicronutrient needs of infants, and necessitated thesecond phase of work that is currently underway.This aims to assess the benefit of α-amylase-treated,multi-micronutrient complementary blend ongrowth and micronutrient status of infants aged 6 –9 months old. The complementary blend, developedin phase one of the study, was industriallyprocessed by extrusion cooking in collaborationwith Quality Feeds Limited. The food was fortifiedusing a multiple micronutrient premix based on thelatest WHO recommendations for infants 6 – 9months old. The fortified blend was either treatedwith α-amylase or not.

Mother-infant pairs, recruited at Chilenje clinic,were randomised when the child attained 6 monthsof age to either receive α-amylase treated fortified

Modification ofComplementaryFoods in Zambia

By Victor Ochieng Owino

Victor is currently a PhD studentat the Centre for InternationalChild Health, University CollegeLondon. A Food Science andTechnology graduate, he haspreviously worked as a researchassistant at the University ofNairobi and held a technicalposition at a Nairobi based foodcompany.

The author wished to acknowledge the contribution of the Ellison Medical Foundation,Nutrition Third World and Score Africa to thiswork.

This article is an overview of an ongoing study inLusaka, Zambia, which is working on modifyingcomplementary (infant) foods with α-amylase. Giventhe critical need to develop at a local level, palatable,affordable and nutritious complementary foods forHIV positive mothers which allow abrupt weaningfollowing breast-feeding, the initiative described hereis an important development. (Ed)

blend, or non-α-amylase treated fortified blend. Thecontrol group comprises mother-infant pairs whowere recruited when the infant was 9 months oldand were measured once, alongside mothers andchildren from the intervention groups.

Each infant in the intervention is provided with2 kg of porridge flour per month, while those in thecontrol group receive at least 4 kg (2 months supply)of porridge after all the measurements are made.Monthly anthropometric measurements (weight,length/height, body circumferences and skinfolds)are performed on both the mother and the infant.Haemoglobin is measured in the infant at 6 monthsof age and at 9 months. Monthly intake of non-breast milk foods is determined by 24h recall. Asample of both control and intervention mothersreceive a dose of deuterium oxide when the infantturns 9 months, and urine samples are collectedover a period of 14 days to determine infants’ breastmilk intake.

ObservationsThe main strength of the project is the fact the

main ingredients (maize, groundnuts, bambaranutand beans) are locally produced in Zambia. Maizeand pounded groundnuts are the most commoningredients used to prepare porridge for infantsfrom as early as 2 months of age1, while beans arecommonly used to make stews. However, the cook-ing of maize-groundnut porridge normally takes 30– 45 minutes. The advantage of the developed blendover habitually used porridges is that it is alreadypre-cooked and takes a maximum of 20 minutes tocook. Additionally, α-amylase is widely used in thebaking industry in Zambia and can be readilyaccessed from local dealers. Minerals and vitaminsfor fortification can easily be sourced regionally.

The main observation has been that the devel-oped blend is widely accepted by mothers whoreport that their children like the porridge. Mothershave also expressed willingness to buy the porridgeif it were to be available in the shops. This showsthat future scale up of this work is feasible. The via-bility of this work also depends on the alreadyestablished collaboration among government,research institutions and the private food industry.

For further information, contact Victor OchiengOwino, Centre for International Child Health,Institute of Child Health, Guilford Street, LondonWC1 1EH, United KingdomEmail: vowino@hotmail.com

1 Current WHO guidelines recommend that complementary foods be introduced in addition to breastmilk from six monthsof age. For guidance and resources on complementary feeding, see http://www.who.int/child-adolescent-health/NUTRITION/complementary.htm

Research

Chilenje clinic staff with the new

complementary food

A T

om

kins,

Zam

bia

, 2005

13

C-Safe, a consortium of non-governmentalorganisations (NGOs) for the southernAfrica food security emergency, hasrecently completed its second year of

implementation of a coordinated developmentaland relief programme in Malawi, Zimbabwe andZambia. Based upon a diverse set of programmingexperiences, C-SAFE has compiled a study on‘Better Practices in C-SAFE targeted food assistancein the context of HIV/AIDS’. The study examinesprogramming through non-medical, as well as med-ical, interventions.

At its inception, the consortium identified fourgroups that would be prioritised for food assistance,namely:

• households affected by chronic illness• pregnant and lactating women• orphans and vulnerable children• malnourished children under five years.

Chronic illness is used as a proxy indicator for AIDSthroughout C-SAFE programming and literature.

The study is broken down into two key pro-gramming areas. The first area investigates targetedfood assistance for non-medical programmes andlooks at factors involved in establishing the pro-gramme, targeting, appropriate rations for thechronically ill, complementary activities and exitstrategies. The second area reviews guidance onlinking targeted food assistance with medical inter-ventions, specifically tuberculosis, Prevention ofMother to Child transmission (PMTCT) andAntiretroviral Therapy (ART).

The following are among the key conclusionsfrom the study.

Targeted food assistance for non-medical programmes

Establishing the programmeTargeted food assistance programming has been

more effective in reaching its goals when integratedinto pre-existing relief and/or developmental pro-grammes and infrastructures, as opposed to estab-lishing an independent pipeline. Integration intopre-existing programming allows C-SAFE food tocomplement existing relief/developmental efforts,and can serve to strengthen current relationshipswith communities and local structures. It can alsohelp to establish new relationships where C-SAFEexpands a NGOs coverage area. Sensitisation ismost effective in reducing duplicity and increasingcollaboration when it reaches a broad audience ofvillage leaders, local health centre staff, local NGOstaff, community committee members, potentialbeneficiaries and non-beneficiaries.

TargetingWhile stigma continues to hinder the transpar-

ent and direct targeting of AIDS-affected house-holds, it does appear to be diminishing, especiallyin the rural areas where C-SAFE works. Targeting inthe context of HIV/AIDS continues to require theuse of refined proxies and other creative approach-es to reach those marginalised by stigma. Staff sen-sitisation and training sessions have been veryinfluential in breaking down stigma-related barriersand giving staff new skills and confidence withwhich to approach communities. The application ofmultiple vulnerability criteria (and weighting of cri-teria) ensures that from those referred, the most vul-nerable within each group are served.

While admission and discharge criteria for foodadjuncts to medical interventions are more easilystandardised, C-SAFE encourages the establishingof both admission and discharge criteria for non-

medical interventions as well. By establishing cleardischarge criteria (based on treatment completion ormeasurable improvement in food security status),individuals and households can, and do, graduatefrom direct food assistance to other food securityand livelihood interventions, where these are avail-able.

Conducting re-verification on a regular basis tore-assess the food security status of vulnerablehouseholds is an integral aspect of graduating ben-eficiaries from targeted food assistance. While safe-ty nets are an essential feature of a self-reliant com-munity, C-SAFEs HIV-positive beneficiaries (whoovercome opportunistic infections and regain pro-ductivity), with the right support at householdlevel, can graduate from a short-term food aid inter-vention to a long period of self-sufficiency and pro-ductivity, where a community-held safety net willnot be required.

Appropriate rations for Chronic Illness (CI)Given the dual crisis of food insecurity and

HIV/AIDS, a comprehensive review of the tradi-tional food basket ration choices, amount, and rela-tive proportion of components is needed. A mini-mum increase of 10% more energy intake is requiredto maintain nutritional status and avoid weight lossin asymptomatic individuals living with HIV.Targeted food assistance often includes individualswith AIDS-related illnesses, whose nutritionalneeds include a minimum of 20% increase in energyintake, with as much as 50% higher protein require-ments.

When chronic illness strikes the primary bread-winner and/or caregiver, other members of thehousehold become susceptible to malnutrition. Inareas affected by both food insecurity and high HIVprevalence, a household ration (which shouldinclude a nutrient dense commodity such as CornSoya Blend) is preferable to individual supplemen-tation.

Both NGOs and donors have failed to devoteadequate time or resources to identifying, develop-ing or sourcing appropriate commodities for chron-ically ill individuals and affected households. This isespecially problematic where constraints exist tousing genetically modified foods.

Complementary activitiesTapping into private (and other donor) funding

has been an effective way of allowing C-SAFE toconduct complementary programming that was notallowable under C-SAFE’s single donor fundingsource. Linking with partners, both local and inter-national, has brought specialisations/expertise invarious sectors and added value to C-SAFE’s target-ed food assistance programming.

Exit StrategiesA phase-over exit strategy will only be effective

when the community is strong enough and motivat-ed to care for itself and its vulnerable households.As C-SAFE approaches its final year, NGO membersstruggle to obtain adequate cash resources to ensurethat community level institutions are sufficientlyprepared and empowered to assume the responsi-bility of caring for the most vulnerable.

Targeted food assistance for medicalinterventions

TuberculosisC-SAFE has identified TB patients undergoing

treatment for food assistance in its coverage areas.Based on stakeholder interviews, linking food pro-gramming with TB treatment has achieved severalsuccesses, including;

• very high TB adherence through the full DOT (Directly Observed Therapy) cycle

• reduced default rates• increased case identification• observed improvement in well-being includ

ing weight gain• improved return to work/productive activity.

Prevention of Mother to ChildTransmission of HIV

An investment in mothers during this time isintended to assist them in delivering a normal birth-weight infant, and to support the production ofbreastmilk through the duration of lactation. ManyHIV positive women are the heads of households,and the survival of other household membersdepends on her well-being.

C-SAFE members are concerned that up to 20%of infants born to HIV-positive women acquireinfection through breastfeeding. In resource poorenvironments, the WHO recommends that HIV-positive mothers practice exclusive breastfeedingduring the first six months of life and discontinue assoon as is feasible. While this advice presents manychallenges, C-SAFE experience has been that moth-ers struggle to implement these recommendations,not only because of inadequate knowledge and sup-port but because of the lack of weaning foods thatfully meet the nutritional requirements of a younginfant.

Antiretrovirals (ARVs)Many people on ARVs in resource-limited set-

tings are not able to follow food and nutritionguidelines due to lack of access to the food required.This can lead to exacerbated drug side effects,reduced drug efficacy, and compromised adherenceto treatment. For the purposes of establishing a foodassistance protocol, C-SAFE categorises people onARVs into the following two categories:

a) Those on ARVs with symptoms and related complications (at the early stages of treatment) who require not only 20-30% additional energy intake, but probably need a nutrient-dense, palatable commodity while appetite and absorption are restored.

b) Those on ARVs without symptoms and related complications (well-established on treatment) who still have increased energy requirement (10%) and require a high-qualitybalanced diet.

Those without symptoms and/or complicationswould normally have responded well to treatmentand other support. While nutritional support is stillindicated in this group, these individuals would notfit in a C-SAFE food assistance programme. The keyassumption here is that this group is essentiallyhealthy and is engaged in, or has the potential toparticipate in, productive livelihood activities.Where applicable, these individuals should bedirected to sustainable livelihoods/food securityinitiatives. On the other hand, those on ARVs withsymptoms and related complications are oftenunable to participate in productive livelihood activ-ities. This group would require immediate foodassistance (in addition to other support).

In conclusion, targeted food assistance has thepotential to fill a significant gap in the provision ofcomprehensive HIV/AIDS care and treatment,especially during the initial stage of ARV therapy.

1 C-SAFE (2004). Targeted Food Assistance in the Context of HIV/AIDS. Better practices in C-SAFE targeted food programming in Malawi, Zambia and Zimbabwe. A study published by the C-SAFE Learning Centre.

Better Practice in Targeted Food AssistanceSummary of published report1

Research

The WHO (2004) estimates that Pnuem-oncystis Carinii (PCP)2 accounts for about20% of cases of severe pneumonia in HIVinfected children and over one third of all

HIV related deaths in infancy. Severe malnutritionmay predispose patients to PCP. PCP can be pre-vented by cotrimoxazole prophylaxis. Despiteexplicit guidelines from WHO/UNAIDS on pro-phylactic cotrimoxazole (see box), and near univer-sal use in developed countries, usage of cotrimoxa-zole for prophylaxis against opportunistic infectionsin children was only 1% in Africa in 20013 . A recentreview, conducted by Action Against Hunger(AAH) in Malawi, has examined the evidence forusing cotrimoxazole as a prophylaxis for HIV posi-tive malnourished children.

Evidence for the use of cotrimoxazoleThere is an increasing body of evidence of the

benefits of cotrimoxazole prophylaxis. Studies ofcotrimoxazole prophylaxis in African adults, and ofthe recent CHAP (Children with HIV AntibioticProphylaxis) trial amongst Zambian children aged1-14 years5 , have shown improved survival in peo-ple with HIV infection. In Cote d’Ivoire6, dailyadministration of cotrimoxazole to adult patientswho were both HIV positive and smear positive toTB significantly lowered, by almost half, the rates ofdeath (46% decrease in risk of mortality, p<0.001)and admission to hospital (43%).

The Zambian CHAP trial has observed an over-all reduction in mortality of 43% in children receiv-ing cotrimoxazole as part of a placebo controlledtrial, and reduced hospital admission by 23%. Thishas led to WHO, UNAIDS and UNICEF modifyingcurrent recommendations for cotrimoxazole pro-phylaxis. Interim recommendations endorse contin-ued cotrimoxazole prophylaxis (due for expert tech-nical review in early 2005).

Amongst African adults, PCP may be a lessimportant opportunistic pathogen than TB andother infectious diseases that feature at an earlierstage of AIDS disease progression7. Cotrimoxa-zole can particularly reduce bacterial disease andmalaria, with potential benefit to uninfectedinfants as well as infants with HIVinfection.

Problems and evidence of contraindica-tions for cotrimoxazole

There are a number of potential/hypotheticaldrawbacks to mass cotrimoxazole prophylaxis,which include

• The efficacy of sulphonamide containing antimalarials could be reduced by mass cotrimoxazole prophylaxis.

• Given its antimalarial activity, cotrimoxazoleprophylaxis could impede the acquisition of natural malaria immunity by infants8.

• Further investigation is required into whether or not extensive use of cotrimoxa-

Cotrimoxazole as a Prophylaxis for HIV Positive Malnourished Children

By Susan Thurstans, AAH

Summary of review1

Susan Thurstans is the HIV adviser with Action AgainstHunger (AAH), based in Malawi

Thanks to Amador Gomez,Technical Advisor, Acción contrael Hambre (ACH) for his inputinto this article.

Current WHO/UNAIDS guidance oncotrimoxazole prophylaxis

The World Health Organisation (WHO) andthe joint United Nations programme on HIVand AIDS (UNAIDS) now recommend that allchildren of HIV positive mothers receive pro-phylactic cotrimoxazole against PCP from 6weeks of age and continue this therapy untilexposure through breast milk ceases-and theinfant is confirmed HIV negative (rarelybefore one year of age).

Cotrimoxazole should be offered to all HIVexposed infants from six weeks of age usingthe following criteria:

• Any child born to an HIV infected woman irrespective of whether the woman received antiretroviral therapy in pregnancy.

• Any child who is identified as being HIV infected within the first year of life by PCR, HIV serology or by clinical diagnosis of HIV infection (according to WHO/national guidelines).

• Children older than 15 months who have had a PCP event, have symptomatic HIV disease, an AIDS defining illness, or have CD4 percentage less than 15.

Provisional WHO/UNAIDS recommendationson the use of cotrimoxazole prophylaxis(trimethoprim and sulfamethoxazole) inadults and children living with HIV/AIDS inAfrica state that cotrimoxazole should beused for prophylaxis in adults and childrenliving with HIV/AIDS in Africa as part of aminimum package of care.

Source: WHO/UNAIDS, (2000). ProvisionalWHO/UNAIDS recommendations on the useof cotrimoxazole prophylaxis in adults andchildren living with HIV/AIDS in Africa4.

zole accelerates resistance to cotrimoxazole and cross resistance to other drugs of some pathogens in the community, and the complexities of any such interactions.

Cotrimoxazole prophylaxis based solely onHIVexposure, without confirmation of HIV infec-tion status, is likely the only option in resource poorsettings and remains a trade off between possiblebenefit to the infant, versus the risk of resistance toantibiotics and antimalarials.

Adaptations based on evidence for andagainst

Infants under six monthsThe current9 guidelines in Malawi for the use of

cotrimoxazole in children, recommend that cotri-moxazole prophylaxis should be administered to allinfants born to HIV infected mothers from 6 weeksuntil 6 months of age (MOH 2004, based on WHOguidelines, 2002).

Infants over six monthsCotrimoxazole prophylaxis for HIV exposed

and HIV infected children beyond age 6 months is aseparate issue. PCP is less common in infants oversix months and young children, and prophylaxis inthis group might lead to substantial, and negative,effects on successful management of malaria andcommon bacterial infections10. However due to diffi-culties with HIV diagnosis in infancy and the con-tributing risk of postpartum transmission of HIV,PCP could present before diagnosis of HIV can beconfirmed. There is, therefore, a rationale for pro-phylaxis to all those who are HIV exposed.

Influence of testing on prophylaxisCD4 testing, ideally used to decide on prophy-

laxis, is likely impractical in resource limited set-tings, while total lymphocyte count has not beenproven as an accurate indicator of immune status inchildren. There is evidence that, in the African con-text, an HIV antibody test could be highly specificfor infection as early as 6 months of age but thisrequires further research11 .

MalnutritionNo studies were identified by the review on the

use of cotrimoxazole prophylaxis in HIV positivemalnourished children. Follow-up anthropometricdata collected in the CHAP trial has not yet beenreported.

8 See footnote 59 At the time of writing this article (March 2005), the Malawi

guidelines were in the process of change which should be in place at time of print (May 2005). The revised guidelines recommend that all HIV exposed and infected children will receive cotrimoxazole indefinitely from six weeks of age, or inthe case of the exposed, until a negative HIV result >18 months of age. Adults will start cotrimoxazole indefinitely if HIV +ve (stage 3 or 4) or CD4 count <500.

10 Graham 2002. Prophylaxis against pneumocystis carinii pneumonia for HIV exposed infants in Africa, Lancet, vol 360,Dec 14, 1966-1968.

11 Since the revised Malawi guidelines (see footnote 8) support treating all exposed infants, testing will not be so much of anissue for this age-group in Malawi.

1 Cotrimoxazole as prophylaxis for HIV positive malnourished children, Action Against Hunger, January 2005, Malawi. By Susan Thurstans. This research was funded by the National AIDS Commission of Malawi

2 Khan (2003). Pnuemonia, Pnuemoncystis Carinii www.emedicine.com/radio/topic534.htm

3 Tindyebwa et al (2004). Improving the quality of care forHIV infected children in Africa through networking. Abstract presented at the 2004 international AIDS conference.

4 These have since been updated since the CHAP trial (ref:footnote 5).

5 Chintu C et al, (2004). Co-trimoxazole as prophylaxis againstopportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomized placebo-controlled trial. Lancet, 364, 1865-1871.

6 Wiktor et al (1999). Efficacy of trimethoprim-sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Cote d’Ivoire:a randomised controlled trial, The Lancet, Vol 353, May 1, 1469-1475.

7 Gill CJ et al (2004). Reconsidering empirical cotrimoxa-zole prophylaxis for infants exposed to HIV infection, Bulletin of the World Health Organisation, April 2004, 82 (4).

Research

Children with kwashiorkor may especiallybenefit from prophylaxis

AAH

, M

alaw

i

14

15

The deadly triangle of interaction amongmalnutrition, infection and poverty haslong been recognised. Nowhere is this inter-action more apparent than in southern

Africa. According to the 2004 UNAIDS Globalreport on the AIDS epidemic, sub-Saharan Africa ishome to only 10% of the world’s population, butalmost two-thirds of all people living with HIV. Thereport also states that 57% of adults infected in theregion are women, and 75% of young people infect-ed are women and girls. This growing trend in the‘feminisation’ of HIV represents an epidemiologicalshift in infection rates and modes of transmission,which has prevention, treatment and policy implica-tions. As the prevalence increases among women,the risk of transmission from mother to child hasbecome an important public health issue and isgrowing in significance as a route of transmission.

Research suggests that without interventions toprevent mother to child transmission, the risk ofmother to child transmission in Zambia is approxi-mately 40%2. Considering the high HIV prevalencein Zambia, without intervention, about 41,000babies annually will acquire HIV, which translatesinto about 112 new infections per day. The govern-

ment of Zambia is rolling out services to preventmother-to-child HIV transmission (PMTCT) acrossthe country, and they are currently providing serv-ices at 83 sites with the cooperation of implementingpartners.

Rationale for providing food and nutritional support through PMTCT programmes

In addition to their primary objective of pre-venting HIV transmission, PMTCT programmesalso offer an opportunity to revitalise the broadersystem of antenatal care into which they are beingintegrated. With HIV testing facilities in place,health care providers have a unique opportunity toprovide services to pregnant HIV positive womenin the early stages of infection, and to reach theirfamilies. One of the main opportunities this earlyintervention provides is to establish positive dietarypractices. Another opportunity is to provide food towomen at a particularly vulnerable period of theirlives. Very little is known about the dynamicsbetween pregnancy, nutritional requirements, andHIV infection. However, HIV infection is known toincrease energy requirements3, and studies haveshown beneficial effects on birth outcomes of HIV+women associated with supplementation of certainmicronutrients4. Although it is conceivable thatunder nourishment may also reduce the efficacy ofanti-retroviral drugs, there have been no studies todate of the relative efficacy of Nevirapine or AZT inpreventing mother to child transmission in under-nourished populations. There is, however, some evi-dence that low birth weight is associated withincreased risk of HIV transmission from mother tochild.

As a pilot initiative, the World Food Programme(WFP) partnered with PMTCT programmes beingimplemented at seven sites by the Government ofZambia, to provide food assistance. There are threemajor goals associated with food support to thisprogramme: (1) to enable women to participate inthe programme, thus supporting the existing inter-vention to prevent transmission, (2) to supportwomen’s nutritional status at a particularly vulner-able period of their lives, and (3) to gather opera-tional and experiential information on the feasibili-ty and acceptability of such a programme whenlinked to ongoing PMTCT services.

WFP field studyAt the beginning of 2005, as part of an intern-

ship at the WFP Zambia field office, the author con-ducted an analysis of the role and impact of food aidin PMTCT programmes.

The analysis included evidence gatheringthrough interviews, questionnaires, direct observa-tion, document review and analysis of collecteddata. This article focuses on one part of the report,relating to the perceptions of beneficiaries, clinicstaff and stakeholders of the role of food aid inPMTCT programming in Zambia.

Perceptions of beneficiaries During clinic visits, beneficiaries were inter-

viewed about their impressions of the impact ofWFP food. Only one of the six beneficiaries inter-viewed was aware of food availability in the pro-gramme before coming to be tested, suggesting that,in this instance, food was not perceived to be anincentive for testing. In some smaller communitiesin Zambia, however, awareness of food ration distri-bution through the programmes is reported to behigher.

WFP has a policy that food should not be pro-vided specifically as an incentive for HIV testing.Providing food to somebody who is food insecureas an incentive for testing can be seen as coercive—an approach that might force a person to disclosetheir status to the community and therefore placethem at risk of stigma and abuse. The risk of abuseassociated with HIV testing remains very real inmany contexts; an abstract presented by Kusimba atthe XV International AIDS Conference entitled‘Community Perception of PMTCT Services: theKenyan Experience’ found that the key barrier asso-ciated with non-use of PMTCT services was fear of

On the groundperceptions of WFP food assistance andPMTCT in ZambiaSummary of a qualitative study1

By Pamela Fergusson

Pamela Fergusson is a dietitian/nutritionist whoseresearch and practice areas include internationalcommunity health, nutrition, food security and HIV.Pamela is currently lecturing in nutrition and dietet-ics in the UK and working on consultancies andresearch in southern Africa.

The author would like to acknowledge the contribu-tions of the WFP Zambia office and WFP headquar-ters to this work.

Cotrimoxazole has long been recommendedas a prophylaxis for HIV positive severely mal-nourished children who are susceptible to seri-ous invasive bacterial diseases and opportunis-tic infections such as PCP, and are prone to bro-ken down or infected mucocutaneous surfacesand skin areas (the latter in kwashiorkor, espe-cially). However, cotrimoxazole is likely to ben-efit those in the earlier rather than advancedstages of HIV disease.

Current Malawian guidelines for the man-agement of severe malnutrition include system-atic antibiotic treatment but do not include cot-rimoxazole – experiences have found it to be nolonger effective against established seriousinfection, resistance to cotrimoxazole is com-mon, and bacteria causing small bowel bacteriaovergrowth are better targeted by amoxicillin12.However, the reviewed evidence suggests that itmay still work as a prophylaxis, even when inef-fective as a treatment but should not replace thesystematic antibiotics use in therapeutic feed-ing.

Discussion In many areas where 3 by 513 is being imple-

mented, children will have limited access toARVs. Although for an individual patient, pro-phylaxis of opportunistic infections does notconfer the survival advantage of antiretroviraltherapy, prophylaxis could have an importantimpact because of its low cost and ease of imple-mentation14.

From the CHAP trial and the Cote d’Ivoirestudy, there is a growing body of evidence thatwould suggest that all severely malnourishedchildren who are found to be HIV positiveshould be on cotrimoxazole prophylaxis, andthe child should remain on cotrimoxazole indef-initely after discharge from therapeutic feeding.However, it remains to be seen whether cotri-moxazole prophylaxis is still indicated shouldthe child gain access to antiretroviral therapy(ART) and show an improvement in their condi-tion.

Recommendations As ACF is a leading organisation in the fight

against severe malnutrition, and evidence hasshown that HIV and severe malnutrition areinextricably linked, they should take the leadinitiative in the integration of HIV care into thetreatment of severe malnutrition.

In resource limited settings, emphasisshould be placed on the demonstrated benefitsof cotrimoxazole prophylaxis.ACF-Internationalshould initiate guidelines for the provision ofcotrimoxazole to HIV positive malnourishedchildren, through integrating HIV counsellingand testing into TFCs and NRUs in high HIVprevalent countries. In order to ensure contin-ued HIV care, links should be established withlocal health structures for continued provisionof this drug and other services for the preven-tion and treatment of HIV related infections andsevere malnutrition. By using TFCs/NRUs as anentry point and strengthening referral systems,services such as prevention of mother to childtransmission (PMTCT), ART provision and com-munity home based care (CHBC) can be utilisedto maximum benefit to help prevent the viciouscycle of HIV infection and severe malnutrition.

For further information, contact AmadorGómez, ACH Technical Director, email: agomez@achesp.org

12 Malawi Ministry of Health (2002). Draft guidelines for themanagement of severe acute malnutrition.

13 Global initiative of the World Health Organization and UNAIDS to provide antiretroviral therapy to 3 million people with HIV/AIDS in developing countries by the endof 2005.

14 See footnote 5

ResearchThe risk of mother

to child transmission is approx 40% in

Zambia

Van

essa

Vic

k /

WFP

16

1 Report: WFP food support of PMTCT programmes in Zambia: A situational analysis of the needs, response and impact.

2 Zambian Ministry of Health Prevention of Mother to Child Transmission (PMTCT) protocol guidelines

3 Nutrient requirements for people living with HIV/AIDS: Report of a Technical Consultation. WHO, 2003.

4 ML Dreyfuss and WW Fawzi, 2002. Micronutrients and verticaltransmission of HIV-1. American Journal of Clinical Nutrition, Vol. 75, No. 6, 959-970, June 2002

testing positive, and the potential consequences ofHIV positive status. The authors recommended thatcommunity sensitisation, service integration, andwomen’s empowerment could help to minimiserisks associated with PMTCT. These communicationstrategies are in place as part of this programme,however the issue of stigma must remain a consid-eration in programme planning.

Measures of food security include food adequa-cy and dietary diversity. Beneficiaries in the PMTCTprogramme said that WFP food was an importantmotivator in attending follow-up appointments.Women also reported that they were eating a greatervariety of food, including more fruits, vegetables,soya beans, meats and eggs, and were eating morefrequently. Although WFP was not distributingfruit, vegetables, meat or eggs, some beneficiariessaid that the receipt of WFP food had enabled themto purchase some of these items themselves. Thismay have facilitated them to be able to take actionon some of the nutritional education that theyreceive as part of services. One woman said, “Wehave been taught about a balanced diet. I make aneffort to eat more green vegetables and beans.”Another woman said, “We know that we have foodin the house. We used to think, if this mealie mealfinished, where are we going to get the food from?”Four of the six women said they felt they had gainedweight as a result of the food rations, and all of thebeneficiaries said they felt healthier because of thefood. This last response could indicate a perceivedincrease in quality of life due to the food. Pursuingfurther assessment of the impact of WFP support onquality of life, using qualitative research methodswith beneficiaries, would be a good direction forfurther research.

Perceptions of PMTCT clinic staffA workshop-based training session of staff

members was carried out, involving nurses andnutritionists working directly with women and fam-ilies involved in the PMTCT programmes atZambian Ministry of Health, and several MTCTclinics in Lusaka. A ‘problem tree’ was developedon perceived problems related to nutrition and foodsecurity faced by women in the PMTCT programme(see figure 1). The participants wrote these problemson cards, and then arranged the cards on the wallwith the problem they felt was most central in themiddle. The staff chose poverty as the central prob-lem. Problems below poverty in the problem treeare seen as root problems, or causes, and problemsabove poverty were branch or effect problems onthe problem tree. Problems clustered around theproblem of poverty were seen as most closely relat-ed to the central problem.

Participating staff were also asked, as part of asecond exercise, to put forward objectives for theWFP food support of PMTCT programme. Theobjectives they suggested were:

• To promote good nutrition to all pregnant women in the PMTCT programme

• To provide knowledge about preparing nutritious food, eg. balanced diet

• To provide knowledge and skills to health workers in PMTCT programme

• To provide food security to all pregnant women in PMTCT programmes

• To help have healthy babies.

A questionnaire on the effects of WFP food sup-port was returned by fifty-five staff members fromWFP supported PMTCT clinics, or clinics wheretraining was provided in preparation for startingfood support. Of these:

• 86.3% felt that the women’s nutritional status improves because of the food

• 86.1% felt that the food encourages women to return to the clinic for follow-up

• 62.5% felt that the women are less likely to sell off assets because they have more food available at home

• 58.4% felt that the food encourages women to come to get tested

• 54.7% felt that women have healthier babies because of the food

• 51.1% felt that women are less vulnerable to engaging in high-risk behaviours because they have more food available at home.

Additionally, the respondents provided theirown ideas of the impact of WFP food. Some of thosecomments included:

“It will help prevent early progression of HIV intoAIDS”

“If the mother is healthy, the chances of transmittingthe HIV virus in utero (while pregnant) to the babywill be minimised”

“Since we always teach them about the importance ofdiet, with food being supplied it will be very easy forthe clients to understand the whole concept”

“Because the people in our programme are vulnerable.Right now we only give a mother PMTCT drugs andforget about her nutritional status in pregnancy andafter delivery now with food we can help the mother alot more.”

“Yes – most people in our community can not affordthree meals a day. At least if they can have soya por-ridge in the morning, then they may be able to havetheir regular meal in the evening.”

“For those who might be single, divorced or widowedand have no source of income, it could help in sustain-ing them, and prevent them from infecting otherswhilst engaging in activities like sex to buy food.”

Categorising clinic staff responses, 9% (4/44points raised) were programme related, 14% (6/44)were HIV mitigation related, and 45% (20/44) werefood and income security related.

The problem tree reflects that, for clinic staff,food and nutrition issues are complex, connectedand far reaching. The questionnaire responsesdemonstrate that staff feel food assistance toPMTCT programmes can have a holistic impact, onclinical as well as social factors.

Perceptions of expert stakeholders In order to access opinions from experts in the

field of PMTCT both locally in Zambia and interna-tionally, a questionnaire was sent to researchers,policy makers, clinical specialists, government offi-cials and relevant UN employees. One of the fouropen-ended questions asked, ‘what impact do youthink WFP food assistance is having/could have on

programmes to prevent motherto child transmission of HIV?’Of the eight responses to thisquestion received, five wererelated to impacts on maternaland child health, specifically onnutritional status and breast-feeding, and three responseswere related to programmingand food security.

The sample of responsesbelow illustrate the broad rangeof impact that expert stakehold-ers feel food assistance couldhave in PMTCT programmes.

“Lots of women are educated asto the importance of breastfeeding

for the child’s health but a low percentage of women areexclusively breastfeeding until 6 months in Zambia, part-ly due to mother’s malnutrition. Food aid for lactatingwomen coming from food insecure families can be animportant factor contributing to promotion of breastfeed-ing.”

“It is hard to convince people to take medicine andengage in health care if their basic needs are not beingtaken care of. When you are really sick with AIDS, yournutritional requirement goes up, if you give them medi-cine and not food it won’t work. Before medicine has tocome food, without food, medicine can only do so much.”

“WFP food assistance supplements the woman'sexisting diet. A satisfactory diet contributes to goodnutritional intake/absorption and satisfactory healththroughout her pregnancy. To further elaborate on thisstatement, WFP food assistance contributes to steadyweight gain throughout her pregnancy, enables properfoetal growth and development and, ensures that shereceives vital micronutrients. In addition, food encour-ages a woman to take medications, i.e. short-course ARV[therapy]. Regular attendance to antenatal clinic ensuresthat she receives awareness and education on prevention,treatment and care of both herself and her infant (regard-less of their HIV status).”

These responses show that stakeholders feel thatthe impact of food is broad, including improvingrates of exclusive breastfeeding, improving healthseeking behaviours amongst HIV infected people(such as intervention programme uptake and reten-tion), as well as improving clinical outcomes formaternal and child health.

ConclusionsThrough the perceptions gathered here from

beneficiaries, staff, and expert stakeholders, we cansee that food aid has a powerful potential for posi-tive impact on prevention of mother to child trans-mission of HIV programming. The impact of foodappears to be spread across the HIV response con-tinuum of prevention, mitigation, treatment andcare. There are potential risks, including food serv-ing as an incentive for testing and the risk of devel-oping dependence on food assistance. These riskscan best be addressed through integration withother programmes and services, including compre-hensive maternal and child health programmes,health education and capacity building projects, andincome generating projects.

One of the stakeholders spoken to in Zambiahad attended the 2004 International HIV conferencein Bangkok. She said she had not realised, until dis-cussing it at the conference, that only a few coun-tries are operating PMTCT programmes with foodassistance. She said, “WFP Zambia is at the forefrontof food support and HIV. WFP is not there yet on aglobal level, and they need to be.” Taking lessonslearnt from food supported PMTCT programmes,and moving forward with increased coverage inhigh prevalence regions, could have a positive, per-vasive, and broad impact on the future of preven-tion of mother to child transmission of HIV. We needto ensure, however, that we make the best use of ourpilot programmes to systematically measure anddocument the impact of food on PMTCT. This willallow us to scale up with programme designs thatare robust, sustainable, and do not have unintendednegative effects, but rather, maximise the potentialfor food improving maternal and child health out-comes.

For further information on this study, contactPamela Fergusson, email: p.fergusson@chester.ac.uk. For more information on WFP PMTCT related pro-grammes, contact Andrew Thorne-Lyman, PublicHealth Nutrition Officer, WFP, email: andrew.thornelyman@wfp.org

F i g u r e 1

Research

17

When approached to be part of ateam to collect material for a spe-cial edition of Field Exchange focus-ing on HIV/AIDS and nutrition/food

security, I was extremely excited and felt itwould be a great learning opportunity. At anENN planning meeting, we decided to onlyattempt focusing on Africa and in particular, theHorn and Southern Africa. Again, we had to nar-row this down to a small number of countriesand try to pull out experiences from a manage-able number of programmes, given a limitedtravel time of approximately six weeks. The ini-tial planning started in late 2004 and involvedcontacting agencies and pencilling in meetingsand potential projects to be visited.

The field trip started with regional meetingsin Johannesburg in early January 2005. Fromhere I moved onto Malawi for a very hectic twoweeks, rounded off by a trip to the airport for anon-existent flight to Zambia. In some ways, anextra two days to tie down and write up findingsfrom the Malawi trip was welcome, but it didmean the Zambian trip was shorter than initial-ly planned. However, with good co-operation, Istill managed to see quite a lot of people andsome interesting programmes. On to Kenya andthough a mere four day trip that included aweek-end, was extremely fortunate to make it towestern Kenya to see a very dynamic localorganisation. This was largely due to the supportof WFP. But, again, we had flight problems. As Irocked up for the Friday evening flight out ofwestern Kenya, I found that it was over-bookedby 29 people and I was rescheduled to the earlymorning flight on Saturday. I finished off mytravels with two weeks in Uganda, again achiev-ing a substantial amount due to all the support

A Nicholson, GOAL

from implementing partners on the ground.

Working as an independent consultantmeant that I missed the usual luxury of anagency picking me up from the airport, havingaccomodation organised for me, and my itiner-ary all planned. Basically, I had to organise myown itinerary within a very tight timeframe.Therefore I spent some time haggling with taxis,trying to book accommodation prior to arrival(some of the lodgings proving a bit dodgy), andactively locating people/agencies in order to setup meetings. However, thanks to a number ofgood friends in the region, my trip was mademuch more successful than it might have been.I really appreciate their support and hospitalityin providing food and accommodation.Furthermore, WFP was extremely supportive interms of planning, organising and taking timeout to travel to the field with me. Many of theimplementing agencies were also extremelyhelpful, very honest and openly discussedissues.

At times, people were a little surprised at myfocus on HIV/AIDS and nutrition/food security.To many it seemed so logical that there shouldbe integrated programming, and that this shouldbe part of a programme approach, that therewas really little need to discuss a rationale.However, in my fieldwork I found these assump-tions not to be the case. In a region with veryhigh levels of HIV prevalence - 10-20% in manycountries - and underlying poverty affectingbetween 40-60% of the population, food securi-ty in many households is currently a majorissue. This places an extra stress on the house-hold, already marginalised by HIV. Furthermore,many households are made even more food

insecure as a result of hosting orphans from HIVaffected families.

Although all the countries visited during thistrip have many similarities, there are also signif-icant differences. Generalisations about pro-gramme approach are, therefore, not wise. Forexample, in Uganda where there has been a cul-ture of awareness and openness around HIV foralmost two decades, prevalence is now down toaround 6.2% and targeting is much easier (stig-ma is reduced but still present). Furthermore,there are more facilities for testing for HIV.Indeed some of the programmes will only admitbeneficiaries if they have documentation toshow they have been tested. This contrasts withother countries visited where testing is not avail-able countrywide (e.g. Malawi) and stigma ismore prominent so that proxy indicators, suchas chronic illness, are used for programmingpurposes.

It is really very difficult to target ‘chronic ill-ness’ and it probably leads to both high levels ofinclusion and exclusion error. In one pro-gramme, where chronic illness was the initial cri-teria, beneficiaries were offered VCT due to thepotential introduction of ARVs. Around 20% ofthese chronically ill tested HIV negative. Somewere upset to know their status, as it meantthey would lose the resources they were receiv-ing. This seems amazing, as they should bedelighted to be HIV negative. This suggestsmany of these people are living in a state ofchronic poverty and need to be supported, evenif not under this particular type of programme.

When visiting programmes, in particularthose with HIV positive and TB affected benefi-

ENN in the FieldMary Corbett isa food securityand nutritionconsultant whovisited southernAfrica on behalfof ENN in early2005.

Editorial

18

ENN in the Field

1 Zachariah R et al, 2002. Moderate to severe malnutrition in patients with TB is a risk factor associated with early death. Trans Royal Soc Trop Med and Hygiene (2002) 96, 291-94

2 Chintu C et al, (2004). Co-trimoxazole as prophylaxis againstopportunistic infections in HIV-infected Zambian children (CHAP): a double-blind randomized placebo-controlled trial. Lancet, 364, 1865-1871.

3 Anglaret X et al, 2003 Pattern of bacterial diseases in a cohort of HIV-1 infected adults receiving cotrimoxazole prophylaxis in Abidjan, Cote d’Ivoire, AIDS 2003,17:575-584

also in place. Projects aimed at strengtheningmarketing are also being looked at by someagencies, particularly those working in ruralareas.

The introduction of ARVs is going to dramat-ically change the dynamics around HIV/AIDS.ARVs are already allowing people to live longerwith AIDS. However, the rollout needs to becarefully handled. In many of these resourcepoor countries where the health infrastructure isextremely limited, adding the ARVs programmesto health care will stretch an already limitedhealth care capacity. Other components of thecare and support to PLWHA need to be in placein order to support compliance with ARVregimes. These components include basic healthcare, good nutrition support and food security. Itis recognised that PLWHA require at least 10%more energy daily, but when sick with oppor-tunistic infections, this requirement is increasedeven further. The opportunistic infections oftenreduce appetite and therefore can lead to rapidweight and weakening of an already weakimmune system. Therefore good nutrition is amajor factor in good health, particularly withthis vulnerable group.

Recent studies in Zambia2 and Cote d’Ivoire3

have shown substantial reduction in mortalityand hospital admissions for PLWHA after theintroduction of cotrimoxazole as a daily prophy-laxis. The objective of introducing cotrimoxazoleis to reduce risk to opportunistic infections. Theoverwhelming evidence has led WHO/UNAIDS toprovisionally recommend that all PLWHA inAfrica should receive prophylactic cotrimoxazoleas part of the minimum package. However thisminimum package, in general, is not beingoffered. Some PLWHA, who are well informedand can afford the drug, buy it themselves.However, even at a cost of about a dollar amonth, this is not affordable for many very poorpeople. Arguments for investing in preventionappear unchallengeable if one weighs up thecost of a prophylaxis against the benefits, i.e.improved well being of clients, fewer illnessesand less weight loss so that PLWHA remain well,and therefore productive, for longer. Also, infect-ed individuals may not require ARV’s until muchlater in their illness, resulting in cost savings.

Where activities are being focussed on threeby five initiatives (to have 3 million people onARV’s by end of 2005), it is extremely importantthat this does not become the priority to thedetriment of all other support to PLWHA andtheir families. It is important to adopt a holisticapproach, supporting nutrition, food securityand health care in conjunction with ARV rollout.

It is essential to keep people well for as long aspossible so that they can support their families.

I had the privilege of meeting families wherea member was HIV positive, who, with the manytypes of support being offered - food aid, spiri-tual/psychosocial support and food securityactivities - were actually “planning for thefuture”. One particular family we visited inUganda, where both the mother and father wereHIV positive, had enough cassava planted for atleast a year and were proudly able to tell us thattheir eldest daughter was preparing to go toteacher training college. Their inspirational ener-gy showed the power of a well targeted pro-gramme, offering a variety of support to keep afamily unit intact and giving people hope for thefuture.

ciaries, there was huge gratification for the foodcomponent. Over and over people said to methat the food had “saved their lives”. They oftentestified to their poor health status prior toreceiving the food, stating that many of themwere now back to living well, had regainedweight and were in relatively good health.Although these testimonies are anecdotal, theyare nonetheless important and need to be docu-mented with other relevant data collection.

In the case of patients with TB, many pres-ent with moderate or severe malnutrition. Astudy completed in Malawi of 1181 TB patientsfound that 80% were HIV positive and that 57%of the cases were malnourished1. Mortality wasclosely associated with severity of malnutrition,with higher rates in the moderate to severegroup (early mortality within 4 weeks). Thishighlights the need for targeted nutrition sup-port for this group. In general, once patientsenroll on TB treatment, they report that theirappetites improve rapidly. If they have little foodin the household, it is a major problem for themand can result in failure to comply with thetreatment. In some programmes, compliancehas been noted to improve with food aid sup-port.

Implementing partners are grappling withproject inclusion criteria throughout the region.Some programmes use externally imposed tar-geting criteria for individuals, while others, par-ticularly in rural areas, are more dependant oncommunity targeting. The rural and urban con-text can be extremely different. However, innearly all programmes, although exit strategiesare recognised as necessary, they are only beingdiscussed (rather than implemented) at present.In general, there is a feeling that there needs tobe some sort of time frame for inclusion in afood assistance programme but at the sametime, there needs to be flexibility, especiallygiven the difference in vulnerability betweenbeneficiaries.

For most implementing partners, the pro-grammes are intended to be comprehensive;addressing immediate needs in the form of foodassistance (safety nets), and then more long-term food security assistance in the form ofincome generation, loans and agriculture inputs.Implementing partners also recognise that thehealth component is essential, some agencieslink in with either the MOH or other healthfocused implementing partners. There is contin-ual learning with many innovative types of pro-grammes being piloted. At the same time, manyof the more traditional skill training pro-grammes, such as tailoring and carpentry, are

Vanessa Vick/WFP

FFW road rehabilitation, Nsanje

A N

ichols

on,

GO

AL

Mary Corbett

19

participate in the study. Children less than 15months were excluded, due to the unreliability ofresults in the absence of Polymerase Chain Reaction(PCR) testing methods. Of those mothersapproached, 145 children were eligible and 143mothers consented to Voluntary Counselling andTesting (VCT) for their children.

ResultsThe preliminary findings from the first part of thestudy, carried out during the dry season, indicate aprevalence of 30% (CI 22%-37%) countrywide.However, there were significant regional differ-ences, with the highest prevalence rates in thesouthern region at 42%, 16% in the central regionand 32% in the northern region. The study is beingrepeated in February 2005 (hunger gap) to provide aseasonal comparison.

Out of the sero-positive sample, 58% were admittedwith marasmus, 30% with kwashiorkor and 12%with marasmic kwashiorkor. Weight for height zscores on admission are presented in figure 2. Ofthose enrolled in the study, 40% of HIV positive hadbeen previously admitted to a NRU, compared with29% of HIV negative children (p=0.18, not signifi-cant). Clinical condition on admission to NRU isshown in figure 3. As can be seen from table 1, the outcome for chil-dren admitted to the NRU who were HIV positive ispoor, with almost half discharged as failing torespond (discharged at below the target weight).This group also had a higher rate of defaulters com-pared to those who were not HIV positive.

ConclusionsAlthough these results are not altogether surprising,it is extremely important to confirm the high HIVprevalence rates amongst this malnourished popu-lation group. Given this, specific nutritional andmedical requirements need to be better addressedwithin the context of NRU and community pro-gramming. However, there remain many unknownswith regard to the treatment of severe acute malnu-trition in conjunction with HIV positive status - par-ticularly in children. Child friendly services are notavailable in many situations to deal with the issuesaround HIV/AIDS and children, particularly in thecontext of HBC programming.

ENN in the Field

In Malawi, chronic malnutrition (stunting) isestimated at around 50%, although recent stud-ies suggest this has increased substantially and,in some districts, is now around 65%. Rates of

underweight are at 25% and acute wasting around6%. During the dry season, admissions to the nutri-tion rehabilitation units (NRUs) average between20-25 children a month. During the hunger gap, thispeaks at 40-60 children, with higher numbers in thesouthern region. Action Against Hunger (AAH)became operational in Malawi in 2002, in responseto the food crisis. Initially the main programmingfocus for AAH was supporting the treatment ofsevere acute malnutrition, targeting the under five’spopulation through the Ministry of Health (MOH)supported NRUs. AAH are now operational in 48NRUs scattered through northern, central andsouthern regions of the country.

Through a collaborative process between the MOH,UNICEF and many NGOs, national guidelines forthe treatment of severe acute malnutrition havebeen developed in line with international guide-lines. These include protocols on nutrition and med-ical care to deal with acute malnutrition. As part ofthe development process, admission criteria werechanged from admission using weight-for-age (ameasurement of chronic malnutrition), to weight-for-height, a measure of acute malnutrition.Training of the NRU health staff was one of the maincomponents of this programme. Another key ele-ment was sensitisation of senior staff at district levelregarding the importance of appropriate treatmentand resources for this vulnerable group. Ongoingsupervision in the NRUs was also a major compo-nent. In a short period, extraordinary gains weremade in Malawi in standardising nutrition guide-lines, due to extensive co-operation among all theplayers and an openness of the MOH towardschanging practice.

Context of HIV/AIDS As a result of the compliance with the new guide-lines, training, availability of special diets (F75 andF100) and availability of medicines, the treatment ofacute malnutrition improved significantly. With theuse of weight for height as admission criteria, onlyacutely malnourished children were admitted, sothat length of stay in the NRU decreased and betterweight gains were achieved. However, as seen fromfigure 1, the mortality rate remained high. A highprevalence of HIV among those admitted to theNRU was suspected as a cause - two small studies,conducted in Malawi, indicated prevalences of18.9%1 and 34%2. Given this, a comprehensive studywas undertaken to determine the prevalence of HIVin children admitted to the NRUs.

MethodThe objectives of the study were:

• To establish the point prevalence of HIV sero- prevalence of children admitted to NRUs in Malawi

• To describe seasonal and regional variations in the proportions of HIV sero-positive children receiving care in NRUs in Malawi

• To describe clinical outcomes of HIV positive and negative children admitted with malnutrition

• To improve linkages with services providing support to PLWHA

The study was conducted in collaboration withAAH, the Malawi College of Medicine, MOHMalawi and UNICEF. It was a two-part study, car-ried out during both the dry and wet seasons inorder to assess seasonal variations in HIV preva-lence.

Twelve sites were identified, four from each region,in addition to the referral hospital from each region,and three rural hospitals. All mothers and childrenin the NRU over a two-week period were asked to

Impact ofHIV/Aids on AcuteMalnutrition in MalawiBy Susan Thurstans, AAH and Mary Corbett, ENN

This article developed from an interview byMary Corbett (ENN) with Susan Thurstans,Action Against Hunger, Malawi.

The authors would like to acknowledge thesupport of AAH and the contributions of theAAH staff in Malawi, the Ministry of Healthand Queen Elizabeth Central Hospital, Malawito the ongoing work.

1 Rogerson et al (2000) conducted a study whereby new admissions to an NRU (Queen Elizabeth Central Hospital)were tested over a two week period. Of these, 18.9% were found to be HIV positive and almost 30% of these children died.

2 Kessler et al (2000), found a HIV prevalence of 34.4%, from a sample of 250 malnourished children. The overallmortality for this study was 28%.

S T

hurs

tans,

Mal

awi

Susan Thurstans is HIV Adviser withAction AgainstHunger, based inMalawi

NRU garden

S T

hurs

tans,

Mal

awi

S T

hurs

tans,

Mal

awi

Educationsession onWorld AidsDay

Staff andbeneficiariesin MatenduNRU

20

ENN in the Field

Weig

ht/

heig

ht

Z s

core

0-0.5-1-1.5-2 -2.5-3-3.5-4-4.5-5

Lower 97%

Upper 95%

reactive non-reactive

Figure 2

HIV Status Cured Discharge as nonrespondent

Death Default

+ve

- ve

% % % % %

40 2.5 7.5 45 5

2 163 3

Lost toFollow up

76

Table 1 Cure rates amongst children enrolled in the study

Figure 3

The uptake of testing was high. However, it appeared that on occasion, health staffover-sensitivity regarding HIV/AIDS testing, particularly in relation to stigmatisa-tion, may have adversely affected uptake of VCT.Following on from this study, AAH plans to try to use the NRU as an entry pointfor the care of children with HIV/AIDS. A referral system is being set up to link theNRU to services providing care and support to people living with HIV, in order toprovide a more holistic form of care. Services will include, VCT, PMTCT for moth-ers, ART (mainly for adults) and opportunistic infections (OI) treatment whereavailable, family planning and community home based care (CHBC). To supportthe role of NRUs, there is a need to;

• train NRU health staff to identify HIV symptoms and provide more appropriate support to VCT

• strengthen linkages with paediatric units and paediatric ARV provision as, presently, some NRUs are separate from paediatric wards, and

• improve NRU kitchen garden programmes so that these better support training in crop diversification and use of herbs, as well as improved food security activities for households of NRU patients.

The current Malawi nutrition guidelines for the treatment of severe acute malnu-trition may need to be revised to address the special needs of children admittedwith HIV/AIDS. AAH are currently conducting research to monitor the responseof children to therapeutic feeding according to sero-status, in order to determine ifthere is a need to modify current guidelines.

Finally, nutrition guidelines relating to HBC and children need to be re-examined,in particular with regard to ‘positive living with HIV/Aids.’ A number of initia-tives are afoot with regard to this.

For further information, contact: Action Against Hunger, PO Box 145, Lilongwe,Malawi, email: aah@globemw.net

Integrated PMTCTServices in a RuralSetting in MalawiBy Gertrude Kara, WFP and Mary Corbett, ENN

The authors would like to acknowl-edge the contributions of Dr.Athanase Kiromera, Medical Director,St. Gabriel’s Hospital, and Miss HildaKamera, Matron, St Gabriel’sHospital, to this article.

Gertrude Kara is the ProgrammeOfficer responsible for HIV/AIDSprogrammes in WFP Malawi. She haswide experience in the area of sexu-al and reproductive health, popula-tion issues, HIV/AIDS, nutrition andfood security.

Figure 1 Mortality rates (%) for all NRUs (Dec 2002 – Jan 2004)

Dec

-02

Jan-0

3

Feb-0

3

Mar

-03

Apr-

03

May

-03

Jun-0

3

Jul-

03

Aug-0

3

Sep

-03

Oct

-03

Nov

-03

Dec

-03

Jan-0

4

Feb-0

4

Mar

-04

Apr-

04

May

-04

Jun-0

4

Jul-

04

0.3

0.25

0.2

0.15

0.1

0.05

0

This article is based on interviews by Mary Corbett with MatronHilda Kamera and Dr Athnase Kiromera, and a project visit byMary, accompanied by Gertrude Kara, WFP.

St Gabriel’s Mission Hospital is situated in Lilongwe district, about45 minutes by vehicle west of Lilongwe, the capital. Although justoff the main road and reasonably close to the city, it is a very rurallocation. The hospital services a catchment area of 15km radius

and a population of 200,000 people who rely mainly on subsistence farm-ing. Maize is the staple food, while tobacco is the main source of cash andis mostly cultivated in big estates where the majority of the poor work astenants. Most of the people have inadequate land and cannot afford agri-cultural inputs and are thus, potentially more food insecure. In Malawi,health care is supported through a combination of facilities run by theMOH (Ministry of Health) and the CHAM (Christian Health Associationof Malawi) - approximately 40% of the health care is primarily supportedby CHAM, working closely with the MOH.

PMTCT in MalawiMother-to-child transmission (MTCT) is the second major mode of trans-mission of HIV in Malawi. Recent estimates show that MTCT accounts for8 to 10 % of all HIV cases in the country. Although 10 to 15% of perinatalHIV infections occur through breast milk, women, particularly from ruralareas, regard breastfeeding as the natural and preferred means of feedingtheir infants due to high poverty levels.

The health staff at St Gabriel’s felt that there was a need to support preg-nant/lactating women living with HIV/AIDS. In particular, the healthpersonnel were concerned at the outcome of mothers diagnosed withHIV/AIDS, especially those with borderline nutrition status, and theimpact this would have on breast milk output. A study in Kenya had indi-cated that reduced nutritional intake leading to catabolism increased thespeed of the disease process - it is clear that the death of a mother substan-tially increases the risk of death of the young infant.

Thus, following discussions with WFP, a proposal was developed with avery specific objective to support HIV-positive pregnant and lactatingwomen and their families with nutrition support. In October 2003, in col-laboration with WFP, a pilot nutrition intervention to supportpregnant/lactating women with HIV/AIDS and their families was com-menced.

InterventionAll pregnant women during antenatal visits were offered VCT (voluntarycounselling and testing). Women tested positive for HIV/AIDS wereadmitted to the PMTCT programme and benefited from the nutritionsupport programme. On a monthly basis, the family received 50kg maize,4 litres oil, 7.5kg pulses and the mother received 9kg Corn Soya Blend(CSB) or Lukini Phala (locally produced fortified blend) to be mixed withone litre of oil, which was specifically targeted for her consumption. Itwas decided that by including a family basket, it might reduce sharing of

21

weight, it has triggered alarm bells as it was feltthere was “something going wrong”, such as oppor-tunistic infections or other problems. Due to thenutritional support and improvement inhealth/nutritional status, many of the women havegone back to their daily activities such as farming.This has many positive implications, in particularimproving food security within the household.

Abrupt weaning from breastfeeding was a majorissue initially, with mothers finding it difficult toaccept and not complying. However, with muchsensitisation and ongoing education, women nowaccept that this is safer for their babies and reducesthe risk of them becoming infected with HIV/AIDS.Following on from this, there are now forty-fiveclients (mothers and husbands) in this programmeon ARV therapy. Initially it was difficult to get thehusbands to go for VCT. However, as they wereattending the clinic to help with taking home thefood, over time they decided to be tested. At thisstage, all the husbands of the positive women in theprogramme have been tested.

DiscussionIt appears that there has been a major change in atti-tude to VCT within this programme area over thelast couple of years. At the last antenatal clinic day,one hundred and twenty new patients were admit-ted to the antenatal programme and of these, 109agreed to be tested for HIV/AIDS, i.e. 91% of allnew admissions. There appears to be a number ofreasons for this. First, in St. Gabriel’s hospital catch-ment area, prevalence of HIV/AIDS (at 4%) is lowerthat in urban areas, and significantly lower than inthe southern region. It may be that the fear of beingpositive has reduced and people are keener to knowtheir status. Also, the communities are working wellin some areas, openly supporting PLWHA withcommunity gardens, among many other activities.Finally, public testimonies have helped reduce thestigma around HIV/AIDS.

In general, the health staff feel the nutritional sup-port to HIV/AIDS women and their families is verybeneficial for many reasons, including improve-ment of nutritional status and well being. However,due to huge work load, the staff have been unable toevaluate their data to be able to support this hypoth-esis. WFP plan to support the staff to address thisissue.

Although abrupt weaning is now more acceptableto HIV positive mothers, there is a major issuearound complementary infant foods. Lukini Phala isthe recommended food for these children (onlyavailable food) apart from foods from the main fam-ily pot. Animal milk, in general, is not available and,if so, is not affordable. There is a concern that thisfood is not nutritiously adequate at the early stage

ENN in the Field

the CSB/Lukini Phala. The mother received thisfood ration for all of the remainder of the pregnan-cy (normally 4-5 months) and then for eighteenmonths after delivery. The rationale for the longlength of support following delivery was based onthe following reasons:

• To encourage compliance, so the health team could monitor the mother at antenatal visits and mother/infant pairs closely at post-natal visits

• To offer medical and psychological support to the mother living with HIV/AIDS

• To facilitate ongoing educational support• To support the mother with abrupt weaning

when the infant reaches six months, so as to reduce risk of the infant becoming HIV infected through prolonged breast feeding (see box)

• To monitor routinely the nutritional status of the mother and infant

• To ensure the infant is fully immunised• To offer VCT on behalf of the infant at

eighteen months of age• To target opportunistically this group for

activities such as group support, income generation (rabbit distribution, kitchen gardens,agriculture) and education on breastfeeding.

ResultsAt present, there are around 150 women who testedpositive for HIV/AIDS in this programme and whoare receiving nutritional support. As the pro-gramme has only been in existence for around fif-teen months, beneficiaries are only now starting tobe discharged from the programme. Due to thehuge burden of work by the health staff, they havebeen unable to analyse data from the previous fif-teen months but, anecdotally, feel that the pro-gramme has had major benefits for these womenand their families. It is felt that most of the womenhave gained weight. For those women who lost

G K

ara,

WFP

, 2004

G K

ara,

WFP

, 2004

Women attendingan antenatal clinic

UN infant feeding recommendationsrelating to HIV/AIDS

HIV-negative mothers and for mothersunaware of their HIV status

Infants should be exclusively breastfed forthe first six months of life to achieve opti-mal growth, development and health.Afterwards, they should receive nutritional-ly adequate and safe complementary foods,while breastfeeding continues for up to twoyears of age or beyond.

Infant-feeding recommendations forHIV-positive women

Given the need to reduce HIV transmissionto infants while at the same time notincreasing their risk of morbidity and mor-tality from other causes, UN guidelinesstate “when replacement feeding is accept-able, feasible, affordable, sustainable andsafe, avoidance of all breastfeeding by HIV-infected mothers is recommended.Otherwise, exclusive breastfeeding is rec-ommended during the first months of life”and should then be discontinued as soon asthe above conditions are met.

Source, and for further details: HIV andinfant feeding: A guide for health-caremanagers and supervisors. UNICEF,UNAIDS, WHO, UNFPA. WHO, 2003.

This and additional resources availableonline, at http://www.who.int/child-ado-lescent-health/

of abrupt weaning and it is felt that a more appro-priate food is required. The health staff debated onthe appropriateness of the use of the locally pro-duced ‘Plumpy nut’ for this age group. There wasalso a concern voiced that stopping the food assis-tance at eighteen months may also create problemsfor some of the beneficiaries, particularly wherethey are particularly food insecure.

Although women appeared to be willing to be test-ed for HIV/AIDS in this community, men were, ingeneral, much more reluctant. Where women werepositive, some men felt they would also be positiveand hence there was no need for them to be tested.As it is a polygamous society, in some cases the menleft the women and moved to the other wife. Withthe possibility of ARV drugs, this trend may changein the future.

In conclusion, the nutritional wellbeing of a womanplays a key role in the overall status of her own andher baby’s health. Food support improves the ener-gy and protein intake of mothers, helps build theirreserves and reduces their vulnerability to oppor-tunistic infections. Integration of a well-targetednutritional support increases mothers’ and babies’access to health services. Food support enhancesmale involvement in the programme and empowerswomen to face their HIV status positively.

For further information, contact: Gertrude Masautso Kara, World Food Programme,PO Box 30571, Capital City, Lilongwe 3, Malawi.Tel: +265 1 774 666, email:Gertrude.kara@wfp.org

HIV/AIDS counselling session

and 60% of Mulanje populations lived in poverty. InMulanje, Oxfam is targeting half the villages in thedistrict while, in Thyolo, fewer villages are yetinvolved as it is a relatively new district for the pro-gramme. Since MSF-Luxembourg (MSF-L) is alsooperational in Thyolo, a working relationship hasbeen developed to avoid duplication and ensure com-prehensive support to the communities.

TargetingAll interventions are targeted through structureswithin the communities, mainly Village DevelopmentCommittes (VDCs). For the safety nets and healthcareprogrammes, Oxfam works with home based care(HBC) structures, where these exist, and where thereare no HBCs, with VDCs at a community level.Through the HBC structure, beneficiaries are identi-fied, prioritising the worse affected. Food aid and caresupport is given, and longterm food security supportis mobilised and supported. Beneficiary status isreviewed on a monthly basis.

ProgrammeAt present, 3,500 households (HH) are receiving foodaid from Oxfam, while an additional 5,228 HH arereceiving food aid from WFP. A family basket of maize50kg, oil, beans and Lukini Phala1 are given on amonthly basis. This element of the programme com-prises approximately 10% of the programme budget.

Oxfam recognises that there is a need to stimulatehousehold and community production. At presentthis is taking place through training, distribution ofseeds for communal gardens, distribution of fertiliserand development of village grain banks linking pro-duction to vulnerable households.

Health care is conducted through the HBC struc-ture. In Thyolo, a Memorandum of Understandingwith MSF-L is being signed to link programmes andprevent overlap. MSF has the capacity to test and treattuberculosis (TB), have facilities for VCT (voluntarycounselling and testing), and have the capacity to sup-port patients with antiretroviral therapy. With goodcollaboration it is hoped that Oxfam will mobilisepeople to access MSF services, while supporting foodsecurity initiatives for these individuals.

ChallengesThere are a number of challenges for this programme.Food insecurity in the region is a longterm problem,and food aid is a costly and not sustainable long-termmeasure.

The challenges faced are a composite of addressingimmediate needs and longer-term food security.However, without addressing both concurrently, itwould be very difficult to make substantial progress.Although food aid has been very well received andappreciated, it is difficult to measure impact of thisintervention. There is a need to develop tools to meas-ure impact of this integrated approach.

Access to markets is a major constraint, and it may benecessary to consider ‘fair trade’ issues while attempt-ing to strengthen markets. Also, linkages betweenHBC and the health infrastructure are very weak.While HIV/AIDS related needs will remain for manyyears, strengthening community mechanisms mayhelp to ensure longer-term support for the chronicallyill.

Future plansIn 2006, Oxfam plans to expand its present pro-gramme and work in another district, (likely BlantyreRural) using the same approach with an expandedbudget. Oxfam also plans to strengthen their work inthe key areas of advocacy, emergency preparednessand mitigation and capacity building within theMinistry of Health.

For further information, contact Oxfam Malawi, P/B213, Blantyre, Malawi.

1 A locally produced fortified blended food

22

ENN in the FieldThis article describes the evolution of Oxfam’s opera-tions in Malawi, and touches on how Oxfam have beeninfluenced by their regional experience of HIV/AIDS.

Evolution of Oxfam’s approach

Prior to 1996, Oxfam provided financial sup-port from their Zambia based operation to amixture of partner implemented programmesin Malawi. In 1996, Oxfam became physically

operational in Malawi, embarking on direct program-ming as well as working through partners, and focus-ing on livelihood support. Key activities included on-farm production, off-farm production, supportingsmall businesses, and social mobilisation in the formof capacity building. In 1999, the programmeapproach changed within the Oxfam global organisa-tion, with the focus more on strategies to change thelives of the poor. Consequently, this meant a shifttowards policy and practice changes that impact onpoverty. Within this broader programme framework,the Malawi country programme focused on food andincome security for the poor, their right to be heard,and gender. In 2000, in response to a greater aware-ness of the impact that HIV/AIDS was having oncommunities they were working with, Oxfam Malawichanged strategy again, this time mainstreamingHIV/AIDS work in all their programmes.

Following on from the 2002 food crisis, there was arealisation that the effect of the countrywide foodinsecurity was particularly marked for those house-holds affected by HIV/AIDS. In 2003, Oxfam decidedthat a new strategy was required specifically to dealwith this problem since despite reaching the end ofthe drought assistance period, it was felt “it was notpossible to pull out as people were too vulnerable”.This article outlines this new programming strategyfor Oxfam in Malawi.

Rationale of current approachUnderpinning the new strategy was the OxfamMalawi team’s feeling that food aid was essential, inthe short term, to save lives. Without food aid, verysick people will not survive. Bridging the gap is alsocritical in order to support more long-term food secu-rity through strengthening food production. Food aidcan break the cycle leading to individual and house-hold crisis. However, strengthening food productionto bolster longer-term food security is particularlychallenging in Malawi, where agriculture is rain-fedand therefore highly dependent on rainfall anddrought.

The need to support seriously ill patients with healthcare and medications is clearly a priority. Prescribingmedicines on an empty stomach is not advisable andcan influence tolerance to medication. Therefore it isimportant to have both a food aid component andmore long-term food security measures to supportpatients’ health.

In order to have a significant impact, these considera-tions argued for an integrated approach to program-ming for the chronically ill and their families throughprovision of health care, meeting immediate foodneeds, and long-term food security initiatives.

Given the vulnerability of the chronically ill, an inte-grated approach to interventions was developed byOxfam which encompassed;

• Safety nets support in the form of food aid and free inputs.

• Medications and health care with advocacy as a key strategy.

• Support to longterm food security.

PopulationOxfam is operational in three rural districts, Mulanje,Thyolo and Phalombe, in the southern region ofMalawi. The government is a key partner in all threedistricts, as are a number of NGOs. According to the1998 population census, the population of Thyolo is458,976 and that of Mulanje is 428,322. The 2001 inte-grated household survey showed that 71% of Thyolo

Interview by Mary Corbett with MrsNellie Nyang’wa, Oxfam

For the past three years, Nellie Nyang’wahas been the Country Programme Managerfor Oxfam in Malawi. Her previous workincludes 2.5 years in development grantsmanagement (Oxfam), and 6 years inmicrofinance (World Vision International).She is also Vice Chairperson of the NGO Board of Malawi.

IntegratedApproach toSupportingChronically Ill in Malawi

Tending to a household

vegetable plot

This article describes the experiences of a school-feedingprogramme supported by the World Food Programmethrough Project Concern International (PCI).

Zambia, with a population of under 10 mil-lion, is a highly urbanised country, witharound 40% of the population living in themain cities. However, poverty remains a

major problem and according to the living condi-tions monitoring survey of 1998, 73% of the popula-tion are considered to be living in poverty. The HIVpandemic, together with other public health issues,has exacerbated an already chronic situation.Currently it is estimated that 16% of the populationaged between 15 and 49 years is HIV positive1.There is a significant gender difference, with muchhigher prevalence rates among females between 15-24 years compared to men of the same age group(ranges of 17-25% for women, compared to 6-10%for men). The impact of HIV on families is substan-tial, with many households affected and a hugeorphan/vulnerable children (OVC) population as aresult. A Food and Agricultural Organisation (FAO)led study in the Southern Province indicated thatone-third of 766 randomly selected households werehosting orphans.

Community schoolsCommunity schooling is a particular Zambianapproach. Due to the lack of sufficient numbers ofgovernment schools for all school-going age chil-dren, communities have set up their own informalbasic schools. Normally, land is donated by a localchief or church (or sometimes rented) and a struc-ture is built. The community schools are run withthe support of local community based, non-govern-mental or religious organisations. Over time, theinfrastructure expands through a combination ofcommunity financial support and outside donorfunding. Most of the community schoolteachers arenot trained and come from the very same communi-ties that initiated the community school. They usewhatever limited building is available (even open

space) and have limited resources (mainly books,desks, blackboards, etc). Some teachers in theseschools are now receiving government salaries,indicating that the government is acknowledgingand supporting this educational system. Many ofthe schools appear to also have support from churchgroups or religious affiliations. The communityschool movement is managed by the ZambiaCommunity Schools Secretariat, a local NGO man-dated by the Government in the management anddevelopment of community schools in Zambia2.

The urban interventionThe UN’s World Food Programme (WFP) first com-menced food assistance to community schools inLusaka district of Lusaka province in January 2003,with Project Concern International (PCI) as theimplementing partner. After six months of success-ful implementation, the project was scaled up and isnow in two additional districts of Lusaka (Kafueand Chongwe). It serves a total of 205 communityschools, as well as nine residential centres that caterfor street children. Over 67,000 children each monthare being supported with school feeding3. Many ofthe children enrolled in these schools are very mar-ginalised, some are street children and a high pro-portion are affected by HIV (an estimated 40% - per-sonal communication).

CurriculumAs many of these children have dropped out of theformal education system (and in some cases werenever in it), older children of around 10-11 years ofage often end up in grade 1. To try to deal with this,community schools follow an accelerated curricu-lum called SPARK 4. The SPARK curriculum allowsfor only four school years or grades, instead of thenormal seven grades typically adhered to in the for-mal education sector. Thus, two years are conflatedinto one grade. There is also a focus on life skills.Uniforms are not mandatory in community schoolsand there are no school fees - thereby reducingobstacles to attendance. However, if students dowell in the community schools, they have an oppor-tunity to attend secondary, and even tertiary, leveleducation.

Major components of the UrbanIntervention

a) School feeding interventionThe school feeding intervention commenced inJanuary 2003 in the aftermath of the food anddrought crisis of 2002. It was felt that urban popula-tions were as much affected by food insecurity asthe rural population, hence the urban nature of thisintervention. The project targets both communityschools and centres that cater for street children.Each school has its own Parents Community SchoolCommittee (PCSC), which is selected by the local

community and is responsible for the overall man-agement and running of the school. Communityschools are therefore considered a community-based organisation (CBO), and provision of foodcommodities to orphans and vulnerable children(OVC) through community schools is, in itself, aself-targeting mechanism. PCI works in partnershipwith these CBOs with the primary aim of increasingtheir capacity to manage school feeding pro-grammes and overall, OVC programmes. The main objectives of the programme are:

• To improve enrolment and attendance levels in target community schools.

• To increase the participation of HIV/AIDS affected households in activities that will improve their health and livelihood security.

• To improve HIV/AIDS-related knowledge and practices amongst the target OVC population and other household beneficiaries.

• To improve the capacity of PCI’s implementing partners (CBOs) to implement and manage the programme.

The impact of the programme (under objective 1) isassessed by monitoring the number of students onthe school enrolment registers and daily attendancerates.

Wet feeding and dry rationsFor most schools, the assistance is in the form of onecooked meal a day, consisting of a wet ration ofHigh Energy Protein Supplement (HEPS) and veg-etable oil5, a locally produced fortified Soya blenddonated by WFP and delivered by PCI andcookedon site by school volunteers6. All Community SchoolCoordinators have been trained in the variousaspects of managing school based feeding pro-grammes ranging from community mobilizationstrategies, food preparation and handling andhygiene issues as well as report writing skills. Aproject orientation guide was developed in early2003 and has been continuously revised to reflectlessons learnt during project implementation.

Initially it was planned that the feeding would be at08.00 a.m., so that children would not attend classeson an empty stomach. In reality, most communityschools tend to feed the children at approx. 10.00a.m., to allow the community volunteers, who arethemselves wives and mothers, to finish their

SchoolFeeding Programmein ZambiaBy Kate Vorley, PCI and Mary Corbett, ENN

Kate Vorley has spent 12 years working in pro-gramme management at national and regionallevels for international NGOs, faith basedorganisations and multi-lateral donors, with aprimary focus on programmes working withOVC (under health care and education).

The authors would like to acknowledge thework of Project Concern International (PCI) asthe implementing agency, and the World FoodProgramme-Zambia Country Office, as maindonor to PCI/WFPs OVC Support Programme.

1 Source: Central Statistics Office, Zambia, 2002.2 Estimated at over 2000 community school nationwide.3 22 school days estimated for each month.4 School, Participation, Access and Relevant Knowledge5 100g of HEPS and 10g of oil is provided for each

child/school day6 A 50-kg bag of grain is provided to the center to facili-

tate lunch and/or supper meal preparation, in addition to the HEPS supplied.

ENN in the Field

Enough energyto dance!

23

ENN in the Fieldmade towards key education indicators, it is worthnoting that project has also been able to facilitate aconsiderable increase in parent’s participation in theeducation of their children in participating commu-nity schools. Anecdotal reports from the partnersindicate a notable increase in community involve-ment in OVC school activities as a result of theUrban Intervention. High attendance levels by par-ents in community meetings to discuss communityschool plans and identify ways of remuneratingteachers at the school have been reported to PCI.Mothers continue to be primarily involved in thewet ration preparation, as well in collecting fire-wood to facilitate the same.

Future plansSchool feeding will continue for a further one tothree years (Jan 2005 to Dec 2007) under WFPs“Protracted Relief and Recovery Operation” pro-gramme (PRRO). The programme will place moreemphasis on ‘recovery’ than on ‘relief’, and focus onmore sustainable measures to support OVCs at thecommunity school level.

It is recognised that analysis enrolment and atten-dance data from all schools is important in order tobetter understand the impact of the programme.Daily attendance and student results have beenanalysed in June 2003 and July 2004. However, dueto limited CBO capacity, these data have not beendisaggregated on the basis of gender until recently.CBOs have now been trained to do this and genderbased data is being gathered on a daily basis at theschool level and submitted to PCI on a monthlybasis.

PCI aims to put in place a more inter-sectoralapproach to school feeding and it is envisaged thatcollaboration with other UN13 donor agencies willbe increased/realised. Also, additional stakeholdersand funding sources will be identified to ensure amulti-sectoral and integrated support programmefor OVC that access education through communityschools.

For further information, contact Kate Vorley, OVCSupport Programme (Urban Intervention)-ProjectCoordinator, P. O. Box32320, Lusaka, ZAMBIA.Tel: 260 1 256735/6/8,email: kate@pcizambia.org.zm

13 In addition to the aforementioned bore holes delivered to 12 of the schools, UNICEF Zambia has also supplied a variety of teaching and learning materials to 100 of the schools whilst New Zealand Aid has provided similar support to one of the community schools in Kafue District

Many schools find it difficult to turn away students,putting strain on the physical school capacity, i.e.space in classrooms, class size, etc., and creating anadditional workload for the teachers which mayadversely affect the quality of teaching and recordkeeping. As a result, in some schools, the two-classrotation (whereby some students come in the morn-ing while others attend in the afternoon) has beenreplaced by three sessions – especially for the juniorclasses. This means that the students have less con-tact time with the teachers in the school.

The reduced contact time and larger class sizes mayhave a serious negative impact on academic results.However, anecdotally, teachers feel that the schoolfeeding has made a big positive difference with areduction in numbers of children dropping out.Also, many of the children no longer come to schoolhungry, so that concentration levels have improved.There are plans to formally measure the impact oflearning achievement on OVC in the near future.

MonitoringAt the onset, monitoring daily attendance at theclassroom level was poor. Strict monitoring andaccounting of the food, record keeping and storecapacity have also been difficult in some schools.However, measurement tools and training havesince been provided by PCI, and attendance data arenow available. PCI continues to sensitise local com-munities on the rationale for the programme andworks in close collaboration with ‘Zambia OpenCommunity Schools (ZOCS)’. This is a locally basedNGO, who provide additional support in monitor-ing the project at the school level in ZOCS’ affiliatedcommunity schools (21 of the 205 communityschools) who are part of the feeding project.

Food ManagementDue to the increased school numbers and rotationsystem for actual classroom sessions, the on-sitecooking needs to be done more than once a day. Fuelfor cooking is often a limitation, with charcoal notavailable or used up quickly (PCI plans to explorepossibilities of an energy-saving cooking ‘jiko’11 toaddress this concern). Also, securing communityvolunteers to do the cooking has occasionally beendifficult, so that community sensitisation continuesto be a top priority for PCI Food Aid Monitors.

There has been some abuse of food items, particu-larly in urbanized settings. In such instances, distri-bution of food commodities to sites (esp. grain) hashad to be halted while PCI works closely with theschool’s PCSC to put effective food control systemsin place.

Since PCI and WFP cannot guarantee the continuedavailability of grain, the project now uses a specifictargeting tool that aids identification of the mostvulnerable households with OVC12.

WaterWater availability has been an issue in some schools,and although UNICEF have supported certainschools with boreholes, lack of ownership of land incertain cases prevented some needy schools frombenefiting from this component. Communities con-tinue to be sensitised to the need for ‘gettinginvolved’ in revenue generating activities that willenable schools to own their own land.In spite of the above, and in addition to the progress

households chores. All children in the target com-munity schools are fed with the wet ration at thecommunity school to avoid stigma.In addition to the wet ration, individual childrenidentified as particularly vulnerable7 also receive amonthly family take home ration, consisting of a50kg bag of grain. This targeting considers both eco-nomic and social status of households - child head-ed households are considered particularly vulnera-ble, while the dependency ratio at the householdlevel is also taken into account.

b) HIV/AIDS componentThe project also has a Behaviour ChangeCommunication (BCC) component and PCI hastrained a group of 20 youth as trainers in theatre fordevelopment (TFD) with technical assistance fromthe Zambia Open University. TFD involves the useof participatory assessment methodologies for qual-itative and quantitative data collection and incorpo-ration of key messages into drama, song and dance.It was used to strengthen the skills of local dramagroups that work with schools and drop in centresto provide effective HIV/AIDS prevention forbehaviour change and communication. The 20 TOTsin theatre for development skills have, in turn,trained a total of 144 OVC and 64 teachers in thethree districts.For BCC activities targeting OVC in the PCI sup-ported residential centre for street children, a draftcurriculum has been developed for HIV/AIDS edu-cation for this profile of OVC. This curriculum iscomplimented by appropriate Theatre forDevelopment materials that were developed in col-laboration with 30 street children (ranging from 9-20years) from three of the residential centres benefit-ing from the PCI/WFP project.

c) School Based Agriculture component-Pilot ProjectA school based agriculture pilot project has beenimplemented in 10 sites8 under a pilot phase. Theaim of the school garden project is to enable sitemanagement to produce or access resources (fromthe garden or sale of produce) to manage their ownschool feeding programmes. The school garden concept has the following objec-tives:

• To build the capacity of teachers to teach agriculture as a classroom subject under the approved Government syllabus.

• Supplement the material resources available to community schools under the feeding programme.

• Income generation in those sites that involve women’s cooperatives (to ensure that child labor is not used to generate income for teachers).

Two teachers from each of the 10 sites were trainedin vegetable production, field crops, animal hus-bandry and fruit production. A training manual wasused developed with support from a University ofZambia Consultant with input from key stakehold-ers, including the Ministry of Agriculture andCooperatives.

All sites have been provided with in-puts to kickstart the pilot in three main agricultural areas - veg-etable, fruit and poultry production as well as pig-gery. Additional training has been provided in theareas of plant protection, safe storage of chemicals,planting of field crops and fruit trees as well asrecord keeping (physical and financial records).

Findings from the pilot areas will assist in develop-ing a realistic implementation plan for the scale upof this component to other participating communityschools/residential centres.

Ongoing challengesAt present, PCI is supplying WFP food to over67,000 children in 214 community schools and cen-tres in three districts. Absolute enrolment9 and atten-dance rates have increased by 26.6% and 40%10

respectively but there remain numerous challenges. Increased workload

7 Under a prescribed targeting criteria developed by both PCI and WFP

8 The selected schools were selected on the basis of a needs assessment but are representative of the general situation in most community schools.

9 Total number of girls and boys enrolled in all schools and benefiting from PCI/WFP’s UI.

10 Based on 25 schools sampled after training in records keeping had been undertaken in Oct 2004.

11 A portable stove that uses charcoal as fuel. With proper use and maintenance, it has been shown to reduce fuel use by 30 to 50%. See Energy and Resources Group at http://socrates.berkeley.edu/erg/index.shtml and case study at http://www.solutions-site.org/cat2_sol60.htm

12 These OVC need to be attending a PCI/WFP supported school to qualify for the grain targeting.

Children (OVC)taking porridgeat school

K V

orley

, Zam

bia

24

25

ENin theField

CRS Dedza team teaching carpentry and building skills

Katie Cuming and Pajuku BahwereValid Int, Malawi

Rodger Mathisen and Mary Corbett

George Luboobi,Concern Worldwide

AMOS Zaidi and Andrew Nicholson,GOAL Malawi

Mary Corbett with a schoolgirl,Plan Int, Uganda

Reach Out staff Uganda

Team working in school feeding in Zambia, PCI/WFP

Mary Corbett with CRS Dedza team in Malawi

Mary Corbett with women's support group in Uganda (Plan Int)

Staff from Matendu NRU, ACF MalawiA member of GOALUganda team

N

Teresa Banda,Nutritionist,MOH Malawi

Education team, Plan Int, Uganda

Anni Fjord, ConcernWorldwide

Skills development team, CRS Dezda, Malawi

Dr Michael O Carroll,Technical Advisor, MOH Malawi

Home based care team, Meeting Point, Uganda

Mary Corbettand PeterPaul Igu,Reach Out,Uganda

Lola Castroand Gertrude

Kara, WFP,Malawi

Dr Kiromera, St Gabriel's Hospital,

Malawi

Susan Thurstans (AAH) and staff from Matendu NRU,Malawi

Mary Corbett with Mary Makokha, Director of ReepKenya, and REEP staff members)Mary Corbett with the coffee plantation support group

(Plan Int, Uganda)

26

South Africa

ZambiaMalawi

Uganda Kenya

Map adapted from ORACLE ThinkQuest© 1996 NGS CARTOGRAPHIC DIVISION

27

Busia district, in Western Kenya, bordersUganda and Lake Victoria and has an esti-mated population of 369,209 (August 1999national census). The district is subdivided

into six divisions, three bordering Uganda, andincludes the town of Busia, situated on one of themain trading routes between Kenya and Uganda.The HIV prevalence rate in Busia is one of the high-est countrywide, with estimates of 15.5% from ante-natal clinics in 2002. This is substantially higherthan the national average tested in antenatal clinics(8.7% for rural and 12.4% for urban populations).There are many factors contributing to this situa-tion, the primary one being that Busia is the site ofheavy border traffic and therefore hosts high con-centrations of a mobile population, who tend to belinked to risky behaviour and increased sexuallytransmitted diseases such as HIV. There is a similar-ly mobile population around the Victoria lakeside.

Origins of REEPREEP (Rural Education and Economic EnhancementProgramme) is a local NGO operating from Butulatown, also in Busia district. REEP’s mission state-ment is “to improve the quality of life for margin-alised groups in rural communities in WesternKenya”. It was set up by a feisty, dynamic local ladycalled Mary Makopkha who originally came fromBusia District but left for some years to further hereducation. With a degree in communications underher belt, she returned to her native area and decidedto start up her own local NGO. She set this organi-sation up over seven years ago and, in her ownwords, reckons that “the first three years were verydifficult and they achieved very little”. However, tothe outside observer, this seems a little overly criti-cal - this period must have been an enormous strug-gle, laying the foundations of a now strong, grassroots level agency, focusing on community partici-pation and working closely with local leaders.Presently, there are eighteen staff working for REEP,and 40 community health workers (CHW) receivingsome financial support.

Community supportREEP is a grass roots organisation, working with

REEP experiences in Western Kenya By Mary Corbett, based on an interviewwith Mary Makopkha, REEP

A communications graduatefrom Dayster University inKenya, Mary Makopkha origi-nally worked as a freelancejournalist for several publica-tions. She founded REEP(Rural Education andEconomic EnhancementProgramme) in 1997 tospearhead the fight againstHIV/AIDS and other issuesaffecting her Butula-Busiacommunity

The authors would like to acknowledge the contri-butions of REEP staff and donor agencies that havesupported the project.

Rural communitywhere HBC is

operating

M Corbett, ENN, 2005

community management committees (CMC), whichhave been set up with total transparency, elected bythe community. Many of the people within thesecommittees are living with HIV/AIDS. The CHWsare the corner stone of the activities, and workunder the supervision of the REEP Home BasedCare (HBC) department. Training of the CHWs isarranged by the REEP office, in collaboration withPathfinder International. A total of 40 CHWs areworking in this programme, initially trained by theMinistry of Health (MOH) and PathfinderInternational, with further ongoing training provid-ed through the MOH. Income generating activitiesare initiated at two levels, firstly at REEP office leveland secondly, at the support group level, with theassistance of the CHWs.

The CHWs are instrumental in supporting the car-ers. As patients are often very sick, depressed andangry, it can be difficult to look after them, “occa-sionally carers will run away”. Therefore it is impor-tant to have a support mechanism for the carer. Atpresent, the CHWs look after around 3,000 personsliving with HIV/AIDS (PLWHA). Many of theseclients are well, but as they become very ill and pos-sibly bedridden, the CHW will step in to help thecarer with the nursing care.

In addition, 18 support groups have been set up forPLWHA. These groups support one another, organ-ise meetings and plan activities. As they have open-ly acknowledged their HIV status, they are fre-quently an important resource in trying to reducestigma within the community.

There are many other ‘off shoot’ activities of thiscommunity based organisation, e.g. supportingorphan children with school fees, advocating forchange in the widow inheritance custom, support-ing families of people dying with HIV/AIDs, in par-ticular protecting their assets (often family membersof the deceased come to claim all the property).

Food aid programmeA baseline study was conducted prior to the foodaid component of the REEP intervention, whichstarted in 2003. The project goal was to ‘improvefood security among vulnerable households, espe-cially those headed by children and old grandpar-ents’. The food aid component aimed to supportPLWHA and their families during crisis periods. Afamily ration was given, with an extra individualration of corn soya blend (CSB) and oil to the personinfected with HIV/AIDS. The initial food was tar-geted at families identified by the CHW and theCMC, but due to resource constraints this hasproved unsustainable. Plans are now in place tolook at food being given for 4-6 months only, focus-ing on the hunger gap period. Although there is nota major focus on antiretroviral therapy within thisprogramme, MSF is also working in the area andmedicines are available free of charge for those iden-tified as requiring them.

Income generationSmall scale income generating activities have beeninitiated, including community gardens and smalllivestock distributions. It is planned that the com-munity gardens will be at village level. In the past,maize has been the main staple grown in the Busia

area, however this does not grow well. Instead, ini-tiatives are in place to plant more cassava, sweetpotato and traditional vegetables, as well as moresorghum and millet. Some women were given asmall sum of money to set up their own businesses,or to support already established business.

ImpactAlthough it is difficult to measure impact, the REEPteam feel that there are far fewer deaths fromHIV/AIDS since the food aid programme has beenintroduced. “As many as 15 people were dying amonth, now some months go by without anydeaths”. Mary Makopkha feels that many peopledied from starvation rather than opportunisticinfections. At an ENN interview with a supportgroup of PLWHA (approx 20 people), there wasvery vocal support for the food aid. They all claimedtheir health had improved substantially, they hadphysically gained quite a lot of weight, and thatthey “felt much better.” Many had gone back towork and all were mobile and looked healthy.Respondents also claimed that they were able to tol-erate drugs better. As some drugs need to be takenon a full stomach, compliance is compromisedwhere households are food insecure. Many in thegroup said that the combination of the co-trimoxo-zole (Septrin) and food assistance had made a majordifference in their well-being.

An impact study (still in draft form) suggests thatthe very poor households that receive the food assis-tance derive a large proportion of their diet fromthis food aid. In a community where 54% of the pop-ulation are in the absolutely poor category (WFPbaseline study), it is difficult to target the mostneedy. However, it appears that by targetingthrough the CMC, the most needy were better tar-geted.

There is close monitoring of the programme - oneCMC was disbanded for abuse of food aid - butdespite strict targeting criteria, there is some redis-tribution of food due to the high levels of poverty.Awareness of the lack of sustainability of long termfood aid assistance has given rise to income generat-ing activities to address chronic food insecurity.

The WFP programme evaluation (draft) of all imple-menting partners suggests that, in general, the pro-gramme is very beneficial. Wasting in children agedless than 5 years amongst participant householdshas been reduced, compared to non-participants.Meetings with the REEP staff and support groupsindicate that a strong grass roots NGO with a homebased care component has contributed to the suc-cess of the programme. In particular, stigma andbehavioural changes may have been more effective-ly addressed, as all the staff are local and well awareof the context of HIV/AIDS within their communi-ty. Further- more, strong community participation atall levels, with substantial female participation, hasled to empowerment of women in the community,and better targeting of food support.

For further information, contact Mary Makopkha,REEP Programme Director, P.O. box 47-50405,Butula, Kenya. Tel: Kenya 0734-643846.

ENN in the Field

28

Kampala, the capital of Uganda, hosts apopulation of around 1.5 million people.Located close to the shore of Lake Victoria,the city is built on a number of hills, giving

it a somewhat unplanned, scattered appearance.The infrastructure struggles with the expandingpopulation, leading to major congestion on theroads and, in areas where electricity is available, fre-quent interruptions in power supply due to everincreasing demands.

Mbuya parish is situated a few kilometres fromthe city centre. Based on the work of a number ofsmall Christian communities, ‘Reach Out’ wasestablished in the parish in May 2001, to supportpeople living with HIV/AIDS (PLWHA) and theirfamilies. Initially, Reach Out activities mainlyinvolved visiting chronically ill persons in theirhomes. Then, in January 2002, a clinic at Our Ladyof Africa Church was established and over the lastthree years, has grown to become a vibrant organi-sation, providing families of PLWHA with a holisticsupport mechanism. Dr Margrethe Juncker, one ofthe founding members, works tirelessly in the careand support of all the beneficiaries, and her energieshave been instrumental in the growth of this project.

Reach Out activitiesReach Out operates the clinic four days a week.

On the fifth day, patients are visited in their homesand training is conducted. Clinic activities includecounselling and testing for HIV/AIDS (same dayresults, to reduce defaulting), clinical examination,and medical support for opportunistic infections.TB treatment and antiretroviral (ARV) support arepart of the service. People can also meet with a

micro-finance team and receive loans for businessplans.

The project expanded substantially in 2004, withclient numbers up by 68.8%, from 860 clients at endof Dec 2003, to 1452 by the end of 2004. During 2004,a total of 140 clients died, 22% within the first monthof admission onto the project, 42% within the first 3months, and over 60% within the first 6 months.These figures suggest many people come for helpvery late and are already extremely ill. Efforts arebeing made at a community level to encourage peo-ple to attend for testing earlier. Given the actual fig-ures enrolled, it is thought that people are movinginto the area so as to benefit from the services of thisprogramme.

Medical careIn March 2004, Reach Out were recognised by

the National TB programme and since then, a totalof 464 patients have been enrolled in the TB pro-gramme. Free ARV treatment is also available forHIV positive clients with a CD4 count of <250. Atthe end of 2004, a total of 545 clients were receivingfree ARV therapy. In order to support compliancefor both TB and ARV treatment, a community net-work has been established to support people, withenrolees supporting one another. The team consistsof two full time doctors, 12 nurses (one is involvedin counselling), additional counsellors, and 230 vol-unteers - 77% of the volunteers are clients livingwith HIV/AIDs who, in turn, support new clients.

Nutrition supportNutritional support, in the form of WFP food

aid, commenced in June 2002, initially throughAVSI1 (implementing partner), and later, through adirect agreement (MOU) between Reach Out andWFP. Since then, the numbers of people receivingfood assistance has substantially increased fromaround 600 to 1,000 beneficiaries. Reach Out hasalso acquired two containers to store a quantity offood, which assists in better planning of food distri-butions. A total of 10 food monitors, an assistant anda clerk support this project. The daily food assis-tance ration consists of 200g maize, 60g pulses, 20goil and 100g corn soya blend (CSB), per person perday.

Admission criteriaNew clients admitted to Reach Out are assessed

on the day of admission. In the event that the clientis seriously ‘food-stressed’, they receive an emer-gency food ration. Following on from this, the FoodAssistance team visit the client in their home. Theyensure the client is living in the locality and com-plete a WFP socio-economic survey questionnaire,

REACH OUT food assistance in UgandaBy Peter Paul Igu, Reach Out andMary Corbett, ENN

Peter Paul Igu has been a full-time volunteerwith the organisation, Reach Out, since Jan2002, and is the food programme co-ordinatorin Uganda.

The authors would like to acknowledge thework and appreciate the support of DrMargrethe Juncker, the WFP and the Reach Outmedical programme staff in Uganda.

which collects information on family size, foodseaten and meal frequency during the week. It alsocollects information on monthly income, source ofincome, and valuable assets owned by the house-hold (HH). Based on this questionnaire, a decision istaken regarding eligibility for the food assistanceproject. Priority is now given to people on TBand/or ARV treatment. At present, around onethousand beneficiaries are receiving food assistanceon a monthly basis.

Exit criteriaAt present there are no exit criteria, with the

exception of beneficiaries that default for a numberof months. This sometimes happens when peoplemove out of the area, possibly to move back home tofamily. Normally three months of default leads toremoval from the project.

MonitoringAt each clinic visit, all clients are weighed by a

nurse/doctor and have counselling if necessary.Initially when clients are registered in the pro-gramme, they are encouraged to attend the clinic ona weekly basis, then fortnightly and, once stable,monthly. Monthly information is collected on allbeneficiaries on the food assistance programme.Table 1 shows summary data from August 2004,based on WFP monthly reporting.

During August 2004, 56.7% of beneficiariesgained weight, 28.9% lost weight, and 14.4% hadstatic weight. Table 1 further profiles those who hadlost/had static weights according to treatment/infection.

Weight Loss Static Weight

% %

15.4% 15.4%

11.5% 7.7%

26.9% 76.9%

46.2%

TB treatment

ARV/TB treatment ARV/TB treatment

ARV treatment ARV treatment

Opportunistic infections

TB treatment

Table 1 Profile of weight loss/static weight, August 2004

1 Associazione Volontari per il Servizio Internazionale

Clients and staffreceiving lunch outsidethe Reach Out clinc

M Corbett ENN, 2005

Weekly Reach Outmeeting held in alocal church

M C

orb

ett

EN

N,

2005

ENN in the Field

29

Meeting Point :Local CBO inUganda

Hoima District is in the mid-westernregion of Uganda, 200kms fromKampala. It borders the districts ofKibaale to the south, Kiboga to the east,

Misindi to the north and Lake Albert and theDemocratic Republic of Congo to the west. The pop-ulation is predominately rural, with a mainstay ofagriculture and fishing. The main food crops aremaize, millet, cassava and to a lesser extent, matokeand rice. The main cash crops are tea and tobacco.The population is largely dependant on subsistencefarming with many households living in poverty.

The HIV/AIDS pandemic in Uganda has had amajor social, economic and health impact on thepopulation over the last 18 years. There has beenlarge loss of life, leading to an estimated 1.7 millionorphaned children. Property grabbing1 has furtherexacerbated the plight of extremely vulnerablewomen and children, in particular. Stigma over theyears has been particularly problematic and difficultto address.

In Uganda, many national non-governmental orcommunity-based organisations (NNGOs /CBOs)have emerged in response to the overwhelmingproblems created by HIV/AIDS in their communi-ties. A good number of these indigenous organisa-tions have grown substantially in strength and havethe capacity to access international funds to fightHIV/AIDS and implement quality activities.However, many NNGO/CBOs lack the organisa-tional capacity to implement activities based onrecognised best practice or to access funds that areavailable to stronger NNGO/CBOs. Through sup-port from international NGOs, small but committedorganisations, often formed from PLWHAs them-

selves, are able to access financial, technical andorganisation support to assist them to implementappropriate and quality interventions. MeetingPoint Hoima is a good example of a grass rootsorganisation, formed by PLWHAs, who has benefit-ed substantially from such support from GOAL,AVSI (Associazione Volontari per il ServizioInternazionale) and the World Food Programme(WFP).

Meeting PointMeeting Point Hoima is a local CBO, founded in

1991 and registered as a CBO in 2002. It started outwith 30 founding members, all people living withHIV/AIDS (PLWHAs), of whom only five remainalive today. Meeting Point now has 1,424 HIV posi-tive members, of whom 82% are women and 18%are men. Between 1991 and 2003, Meeting Pointwere predominately involved in informal groupcounselling support to members, home visiting thechronically ill, orphan support and medical carethrough AVSI. Its activities have substantiallyincreased over the last few years, which now extendover the whole district

PartnershipsStrong supportive partnerships with AVSI, and

GOAL have contributed to a more comprehensivesupport package for people living with HIV/AIDS.

By Fiona Mitchell, GOAL, and MaryCorbett, ENN

Fiona Mitchell is the Development ProgrammeCoordinator, GOAL Uganda

The authors would like to acknowledge the con-tributions of Veronica, the whole Meeting Pointteam and members visited during the ENN fieldtrip, Stefano Antonetti, AVSI, and Jackie Katanaand Annet Achago, GOAL Uganda.

A Meeting Point staff member with a

young child

M C

orb

ett

EN

N,

2005

Plans regarding targetingAt the end of 2004, the WFP implementing

partners (IP) convened to review the presentprogramme. A small working group was set upto consider admission criteria. Existing ques-tionnaires used by different IP’s were reviewedand a new questionnaire has just been complet-ed and will be piloted in March 2005. This ques-tionnaire has a ‘score weighting system’ which itis hoped will facilitate targeting of the most vul-nerable members in a programme and reducesubjective decision-making. Instead of a cut-offbased on the amount of food available, the cut-off will be based on the scoring system, which isbased on a more comprehensive economic sta-tus appraisal.

However, as yet, no exit criteria have beendeveloped. In general, beneficiaries areinformed that the food assistance will cease atsome stage, although, so far, some beneficiarieshave been on food assistance for 2.5 years.

Income generating activities (IGA) IGAs take a number of forms, the largest

being the ‘Bread for Life’ micro finance pro-gramme. This activity has seen a four-foldincrease from 154 loans in 2003, to 650 loans in2004. Clients must submit a business plan beforea loan is sanctioned by the committee. The sizeof the loans has increased, now up to 100,000Ush ($60) and repayment takes place over a sixmonth period. The average recollection of loansis 81%. In general, 75% pay well, 15% are slowpayers and a further 10% don’t pay. The mainactivities in business plans include vegetablepurchase and sale, stone quarrying, fishshelling, purchase and sale of second handclothes, hair dressing and brewing. Clients mayreceive new loans once the old one has beenpaid off. Clients need to be registered in the pro-gramme for at least three months and attendingclinics regularly before loans will be sanctioned.

ConclusionsIt is felt that most people who attend Reach

Out are poor or very poor. Initially there wassome stigma around food assistance but aftersome time, it appears that food acted as anincentive for people to come to be tested forHIV/AIDS and receive assistance. For TBpatients, in particular, once on medication,appetite increases and food assistance becomesessential to support well-being and compliance.Also, staff feel that TB patients are now recover-ing much better since the introduction of foodassistance, and describe how “less TB patientswere dying”. Similarly, it is considered essentialfor clients on ARVs to receive food assistance sothat their general condition improves. A oneyear time frame for food assistance to clients onARVs may be introduced, which, it is felt,should be flexible.

Reach Out has a significant amount of datacollected on clients over the years, i.e. weighttrends, morbidity and mortality rates.Unfortunately, due to the enormous workloadof day-to-day project implementation, it hasbeen impossible to analyse these data. Projectstaff would welcome both technical and finan-cial support in order to make use of these data,which could be invaluable in answering some ofthe key questions around HIV/AIDS and nutri-tion support/food security.

The volunteers in the programme, many ofwhom are HIV positive themselves, are inspira-tional in their positive attitude to life and theirillness. In conjunction with the enormous com-mitment from other staff, this is probably one ofthe major factors contributing to the success ofthis programme.

For further information, contact Peter PaulIgu, P.O. box 6562 Kampala, Uganda. email:Igupeter@yahoo.com, tel (Uganda) 077-343027

ENN in the Field

GOAL

Capacity building,

Home BasedCare,Positive

livinginitiatives,Sustainablelivelihood initiatives,Advocacy(rights ofPLWHA andOVCs)

MEETINGPOINT,HOIMA

AVSI

Formal/non formaleducation,Food assistancewith WFP, Home Basedcare,Capacitybuilding,

ARV’s withMOH

extremely food insecure.

Case Study 3Moses, in his mid to late fifties, is married with

five children. Both he and his wife, Ester, are HIVpositive. Since August 2004, Moses has been takingARVs. Ester’s CD4 count remains high and there-fore she does not require ARVs at present. Despitethe relatively low cost of antibiotic prophylaxis, ataround $1 a month, Ester is not taking Septrin. Since2003, Moses and his family have been receivingWFP food rations. Through the Meeting Point sus-tainable livelihood initiative supported by GOAL,he has been growing more than enough cassava,beans and maize to feed his family.

Moses became a member of Meeting PointHoima in 1999. Following CCA training in 2003,Moses now heads a team of motivated CCA’s in hislocal area that provide HBC activities to PLWHA’s.The main activities encompassed in HBC includepsychosocial and practical support to families strug-gling with the impact of HIV/AIDS on their lives.Although Moses is on ARVs, he is an extremelyactive member of the community. Through healthcare support, food aid assistance and sustainablelivelihood support, Moses and his family are cur-rently food secure. Due to this comprehensive pack-age of support, Moses and his family are optimisticabout their future and hope to support their eldestdaughter through teacher training in the next fewmonths.

Case Study 4Sarah, in her late twenties, was deserted by her

husband within the last year. She remains in thefamily home close to her own family with her threechildren. Sarah and her husband are both HIV posi-tive. They live in a close knit rural community. Itwould appear that her husband left due to fear ofHIV/AIDS stigma from the neighbours and he nowlives in town with a new partner. He supports thefamily financially on an irregular basis. Sarah is justabout to complete a full course of TB treatment andwill start ARVs in the very near future. Throughouther TB treatment, Sarah did not receive any food aidsupport, and was very dependant from day-to-dayon food support from her family. At present, Sarah’shealth remains fragile and she is not well enough toprepare her garden for the coming planting season.Without any active adult members in the house-hold, she and her three children will remain foodinsecure for at least the next year.

Lessons learnedLesson 1It is evident that women and children’s vulnera-bility increases when HIV affects the family unitand that vulnerability increases with poverty. It isnot uncommon for women in such situations tobe abandoned with children to support,divorced, widowed and disinherited.

Lesson 2Food assistance supports the immediate needs offragile families.

Lesson 3Health care support such as TB treatment, pro-phylaxis for opportunistic infections, PMTCTand psychosocial support through HBC, are nec-essary along with food aid to support immediaterecovery.

Lesson 4In order to achieve more longer term food securi-ty at household level, it is essential to incorporateappropriate sustainable livelihood initiativesalong with food aid.

Lesson 5Nutrition training through HBC increases theknowledge around the use of locally availablefoods. Changing traditional cooking practices,particularly of vegetables, can increase the nutri-tional value of food and helps PLWHAs achievea more balanced diet.

Lesson 6A family ration would be more affective in sup-

Nutrition support and food securityWFP food assistance support to PLWHAs first

became available in Hoima through a partnershipwith AVSI in April 2003. Meeting Point and AVSIwork in partnership to identify beneficiaries anddistribute monthly food rations to 500 PLWHA’sand 50 mothers on PMTCT. The initial criteria forinclusion in the food aid programme was member-ship of Meeting Point (i.e. being HIV positive).However, as time went on, the criteria were modi-fied to target the most vulnerable households with-in this group.

WFP is soon introducing an eligibility format.This tool will provide the possibility to phase outcurrent beneficiaries whose household has reachedfood security other than introducing criteria ofselection. Moreover, when a primary beneficiarydies, the remaining family members will receivefood for a further three months and then will be dis-charged from the programme.

To complement the existing food aid pro-gramme and increase longer term food security,GOAL, in partnership with Meeting Point, fundssustainable livelihood interventions, such as small-scale agriculture/animal initiatives. This is in theform of agricultural inputs and technical support to150 families of the 500 beneficiaries currently receiv-ing food assistance through AVSI and WFP.

Case studiesDuring a field visit in February 2005, four case

studies of Meeting Point members were compiled,highlighting the unique problems created byHIV/AIDs, poverty, stigma, and in particular,female vulnerability.

Case Study 1Mary, in her mid to late fifties, is a grandmother

and a member of Meeting Point since 1992. In 1989,Mary’s husband died, leaving Mary and three otherwives widowed. At this time, Mary lost all her prop-erty and was forced to move with her four childrento Hoima town, where she resided with a friend andgenerated an income by selling tea. Through psy-chosocial support, food assistance and sustainablelivelihood interventions, Mary is now food secure,living well, and in her own home with income gen-erating from a variety of crops/vegetables growingin her garden, pigs and chickens. In 2003, Marybecame a Community Counselling Aide (CCA) withMeeting Point Hoima where she supports otherPLWHA’s through her HBC visits. Mary pays forher own prophylaxis for opportunistic infections(Septrin), which costs about $1 a month, and doesnot require antiretroviral medication (ARVs) at thistime. Through the CCA training Mary received, sheexplained that she is now better able to prepare anduse local available foods to achieve a well balanceddiet.

Through the combination of psychosocial sup-port, food aid, nutrition training and sustainablelivelihood initiatives Mary is now food secure, aproductive member of her family and communityand is presently very well.

Case Study 2Rose, in her late twenties, was widowed recent-

ly and has four children. Due to lack of supportfrom either family and for economic reasons, Rosechose to remarry. Since the death of her first hus-band, one of her children has died, two now livewith her dead husband’s family and the youngestremains with Rose and her new husband. Sinceremarrying and becoming pregnant, Rose joined theMeeting Point/AVSI Prevention of Mother to ChildTransmission (PMTCT) programme in 2004 whereshe has been receiving a single persons food ration.She now has a one-month old baby boy. Since deliv-ery, Rose was sick and admitted to hospital with afever. Rose was distressed at not having the funds topurchase milk for her baby, yet he is a healthy andwell nourished baby (on breast milk only). She wasunaware of current best practices, which promotesexclusive breast-feeding and abrupt weaning at 6months of age. Despite support through the PMTCTprogramme, Rose and her new family remains

1 In patriarchal societies like Uganda, when a man dies, his land typically goes to his male children or to his male kin, reverting back to his clan. Though illegal, property grabbing -when a man's relatives descend upon his widow to claim the household's material possessions - is common.

2 See research review, Cotrimoxazole as a prophylaxis for HIV positive malnourished children, in this issue of Field Exchange

porting a mother through the PMTCT pro-gramme since it is expected that any ration takeninto the home will be shared.

Lesson 7Clear information for mothers in PMTCT regard-ing current best practice for infant feeding isessential.

Lesson 8There is strong evidence to suggest that prophy-laxis for the reduction of opportunistic infections,in combination with nutrition support - immedi-ate and longer term (food aid and sustainablelivelihood interventions) - help to keep PLWHA’swell, for longer2. This may mean CD4 countremains high and progress on to ARVs isdelayed.

Lesson 9Even though prophylaxis for opportunistic infec-tions could be considered a cost effective inter-vention (approx. $1 per month per beneficiary), itis not currently part of any HIV/AIDS interven-tion in Hoima.

For further information, contact Fiona Mitchell,email: dpc@goaluganda.com or Mary Corbett, email: corbettmary@eircom.net

A signdirectingpeople tothe localcounsellor

M Corbett ENN, 2005

M C

orb

ett

EN

N,

2005

30

A mother and childinvolved in Home

Based Care

ENN in the Field

News

31

UNHCR and WFP have developeddraft practical guidance on imple-menting programmes that incor-porate both HIV/AIDS and food

security/nutrition activities. The material istargeted at managerial and technical staff ofthe UN and cooperating partner agencies.The premise of the guidance is that asrefugees often depend, at least in part, onexternal assistance for nutritional, healthand other basic needs, refugee settings pro-vide unique opportunities to implement tai-lored interventions that mitigate the effectsof the illness and prevent HIV transmission– thus addressing the vicious cycle ofHIV/AIDs and food insecurity.

The document describes two types ofintegrated programme strategies. Strategiesdescribed in Chapter 3 incorporateHIV/AIDS prevention, care, treatment andsupport activities into food and nutritionprogrammes. Strategies presented inChapter 4 incorporate food and nutritiontraining, or use food resources for capacity-building and/or institutional support activ-ities in HIV/AIDS programmes. The docu-ment clearly states that it is not always truethat programme strategies across sectorsshould always be multi-sectoral. Food andnutrition programmes are not a ‘magic bul-let’ for preventing HIV transmission or mit-igating the effects of the HIV/AIDS pan-demic. Additionally, it is vital to avoid over-reliance on food programmes, which coulddisrupt agricultural production and marketsin refugee communities. Agency staff shouldidentify interagency and inter-sectoral link-ages that make sense in the local context.

The draft guidance documents state thatthe decision to implement one or more ofthese integrated programme strategiesshould follow from a local assessment,preferably conducted in both the refugeeand host communities. Across refugee set-tings, the epidemiology of HIV/AIDS varieswidely, with some populations exhibiting alow HIV prevalence, except in defined high-risk groups, while others struggle with amore generalised epidemic. HIV epidemicsmay also be associated with different pat-terns of food and nutritional insecurityamong vulnerable groups. Assessment andanalysis of food security and malnutrition inaffected populations are essential in decid-ing whether food-based programmes areappropriate in the context of HIV/AIDS.Initial HIV prevalence rates and the natureof refugee-host interactions can be signifi-cant determinants of HIV risk in both popu-lations. The document describes manyexamples of agency field staff drawing onrefugee and host community skills (e.g.music, dance and drama, teaching, trainingand peer education skills) to build morecommunity-driven refugee health pro-grammes to prevent HIV transmission andprovide medical, health, nutrition and liveli-hood support to the HIV/AIDS-affected.

Finally, the document stresses thatimplementation of integrated programmestrategies should be accompanied by rigor-ous monitoring and evaluation of bothprocess and outcome indicators. Pilot-test-ing and evaluation will provide valuabledata on the effectiveness of these activities.

For further information contact Paul Spiegelat email: SPIEGEL@unhcr.ch

UN DraftGuidance onProgramming

CRS held a Conference on HIV/AIDSand Food Security in Southern Africafrom September 22nd-29th inJohannesburg, South Africa. The aim

of the conference was to appreciate the progressto date, to generate more innovative and long-term HIV/AIDS and food security strategies,and to further increase the quality and scale ofholistic and sustainable programmes thataddress food security, HIV/AIDS and social jus-tice. The conference was a collaborative effortbetween the CRS Southern Africa RegionalOffice (SARO) and CRS headquartersProgramme Quality Support Department(PQSD): HIV/AIDS Unit. The conference wasmade possible through the financial support ofthe Food For Peace grant, the ‘InstitutionalCapacity Building’ (ICB) grant (AFP-A-00-03-00015-00).

Thirty-nine CRS staff attended the confer-ence – 24 from country programmes (includingthree country representatives and three heads ofprogramming), seven from SARO, five fromPQSD, one from OSD, and two advisors fromother regions in Africa. In addition, 18 resourcepeople representing CRS, donors, other NGOs,and people living with HIV and AIDS also con-tributed at the conference.

The conference began by taking stock of theimpressive amounts of high-quality program-ming integrating HIV/AIDS and food security

CRS Conference in South Africa

The second edition of the Food and NutritionTechnical Assistance (FANTA) publication,HIV/AIDS: A Guide For Nutritional Care andSupport1 , is now available, made possible

through the support of the FANTA Project by theOffice of Health, Infectious Disease and Nutritionand the Office of HIV/AIDS of the Bureau forGlobal Health at the US Agency for InternationalDevelopment.

This guide provides information for HIV/AIDSaffected households and communities on how nutri-tion can help HIV-positive people live healthierlives throughout the progression of HIV disease.The guide is an interpretation of the best availableevidence to date from multiple sources, especiallythe World Health Organisation (WHO) and a broadrange of experts. This guide was extensively revisedfrom its earlier version and was peer reviewed.

The purpose of this guide is to assist pro-gramme managers and health workers make recom-mendations on food management and nutritionalissues for households with members who are HIV-infected or living with AIDS, and for individuals,families, and communities affected by HIV. Foodand nutrition recommendations are for both adultsand children and emphasise the use of locally avail-able food products, complemented by appropriatefoods obtained through external donation pro-grammes.

This guide is targeted primarily at four types ofaudience:

1. Programme managers and technical staff who work in food aid, food security, health, and nutrition programmes in HIV-affected areas

HIV/AIDS: A Guide for Nutritional Care and Support

currently being implemented by CRS countryprogrammes. Country programme presenta-tions were made in the areas of Orphans &Vulnerable Children (OVC), Title II food aid,and on the use of frameworks in project devel-opment.

New frameworks and information wereintroduced to deepen country programmeunderstanding of the linkages betweenHIV/AIDS and Nutrition, Livelihoods andSocial Justice. A full day of in-depth trainingwas conducted in each of these thematic areas.With the benefit of this new information andtraining, participants analysed a series of criticalissues such as prevention, complex programdesign issues, new programming areas anddonor resource allocation strategies.

Finally, participants established capacitybuilding and learning agenda priorities for link-ing HIV/AIDS and food security in the comingyears. By supporting the continued programmegrowth warranted by the humanitarian context,while at the same time deepening programmequality, it is hoped that this learning agenda willcontribute to greater visibility of CRS leadershipin this important field.

A report of the conference can be obtainedfrom Kristin Weinhaur email:kweinhau@catholi-crelief.org

2. Local health workers in areas affected by HIV/AIDS

3. Community-based organisations working in high prevalence HIV/AIDS areas

4. Institutions caring for PLWHAs or orphans and other vulnerable children infected or affected by HIV/AIDS.

Six chapters deal with thematic issues related toHIV/AIDS, which are:

• Nutrition and HIV/AIDS: Basic Facts • Managing HIV Disease Through Nutrition

Interventions • Nutritional Issues Associated With Modern

and Traditional Therapies • Nutritional Care and Support for Pregnant

and Lactating Women and Adolescent Girls • Nutrition and Care Recommendations for

Infants and Children • A Food-based Approach to Support

HIV/AIDS-affected Households and Communities

Copies of the guide can be obtained from: Food and Nutrition Technical Assistance (FANTA)Project, Academy for Educational Development,1875 Connecticut Avenue, N.W. Washington, D.C.20009-5721. Tel: 202-884-8000 Fax: 202-884-8432Email: fanta@aed.org. The guide is also availableonline, http://www.fantaproject.org

1 HIV/AIDS: A Guide For Nutritional Care and Support. 2ndEdition. Food and Nutrition Technical Assistance Project,Academy for Educational Development, Washington DC, 2004.Published October 2004

32

Anew study on communitybased technology change tocombat HIV/AIDS in Africais underway, funded by the

MacArthur Foundation. This project isbeing led by staff at Tulane University,with input from the OverseasDevelopment Institiute (ODI). The proj-ect seeks to learn about community-based efforts to adapt technologies(tools, devices, and systems of technicalknowledge) to respond to the impacts ofAIDS throughout rural sub-SaharanAfrica. Eventually this will lead to rec-ommendations for aid agency, NGO andgovernmental policies to support appro-priate technology innovation, adapta-tion and dissemination to combatHIV/AIDS.

Right now, the research team areengaged in a ‘scooping’ exercise to iden-tify a range of local, small-scale technol-ogy innovations and adaptations thatmerit more attention. Examples mightinclude conservation agriculture tech-niques, preservation and processing oflocal plants for enhanced nutrition, andtext-messaging for enhanced care ofpeople living with HIV/AIDS(PLWHA). Also of interest are new tech-nologies for income-generation activitiesfor PLWHA and AIDS-affected house-holds, and innovations in householdwater, energy, and transport. This proj-ect will conduct fieldwork in Easternand Southern Africa to try to find outmore about these innovative technolo-gies at a local level. Many agency reportsmention what external agents should dobased on first principles and/or what isneeded (‘labour-saving’ devices). Lesscommon is news of adaptations andinventions by the blacksmiths, healers,farmers, rural women, youth, orphans,and other community actors and organi-sations who directly, and daily, face theepidemic and its cumulative burdens.

The researchers are requesting infor-mation on specific examples that couldbe investigated further. This might be inthe form of reports, internal assessmentsor anecdotal stories. Information on keyinformants or essential places for theresearch team to visit would also beappreciated.

If you have further information for theresearch team or would like furtherinformation regarding this study, pleasecontact:Paul Harvey, Research Fellow,Humanitarian Policy Group, OverseasDevelopment Institute, 111 WestminsterBridge Road, London, SE1 7JD, Tel (UK): + 44 (0) 20 7922 0374, email: p.harvey@odi.org.uk

CommunityBasedTechnology toCombatHIV/AIDSAn open invite is extended by Ted

Greiner to join The World Alliance forBreast-feeding Action (WABA) HIV andInfant Feeding Yahoo Group.

The group is not a chat group, but a listmaintained by Ted Greiner, in his capaci-ty as coordinator for the WABA ResearchTask Force. Members have access to a pri-vate website with numerous links to keyUN and other documents on infant feed-ing and HIV, as well as unpublished andrecently published full-text journal arti-cles.

Anyone who wishes to join should email TedGreiner email: tedgreiner@yahoo.com

Fact Sheets onARV Treatmentfor NGO/CBOStaff

The International HIV/AIDS Alliance isdeveloping a set of fact sheets and partici-patory tools to support community engage-ment for antiretroviral (ARV) treatment.

The materials are intended for use by non-govern-mental/community based organisations (NGOs/CBOs) , training organisations and individual train-ers, international NGOs and NGO support pro-grammes, to enable staff to support PLWHA andtheir communities on ARV treatment. They arebased on experience in supporting treatment pro-grammes in several countries.

Information covered some so far by the factsheets include:

• adherence to ARV treatment

Aseries of publications have recently beendeveloped to support those working withPLWHA, produced by the Regional Centrefor Quality of Health Care (RCQHC),

Kampala, Uganda. Technical input by Food andNutrition Technical Assistance (FANTA) andLinkages projects were financially supported byREDSO/ESA (Regional Economic DevelopmentService Office/East and Southern Africa) and USAID.

The series comprises of two booklets that accompanya series of counselling cards as a ready-made flipchart:

• Food and Nutrition Counselling for PLWHAon Antiretroviral Therapy - A job aid for counsellors and anti-retroviral therapy (ART) service providers

• Nutritional Care for People Living with HIV/AIDS – Answers to frequently asked questions

Local Resources for Supporting PLWHA

Invite to JoinWABA HIV andInfant FeedingGroup Five new Key Issue guides are now available at

the HRC/Eldis* HIV and AIDS Resource Guidewebsite. Topics, which include nutrition, AIDScommunications and sexual and reproductive

health, have been produced in collaboration with sub-ject experts. These guides provide in-depth coverage ofHIV and AIDS, highlighting important research andlinking to summaries of documents, websites and otherresources. All documents are open access.

To access these pages and more, visit:http://www.eldis.org/hivaids/keyissues.hmHIV/AIDS and nutrition can be directly accessedat:http://www.eldis.org/hivaids/aidsnutrition.htm

*The Health Resource Centre (HRC) provides informationsupport to the UK Department for International Developmentand its partners on public health and communicable diseases inlow and middle income countries. Eldis is an electronic gatewayto development information, hosted by the Institute forDevelopment Studies, UK.

New e -Resourceson HIV and AIDS

• Nutrition for PLWHA – Counselling cards

The job aid booklet provides step-by-step infor-mation to help counsellors and service providers toeffectively counsel on nutrition and HIV for PLWHAson ARVs. The FAQ booklet has answers to commonlyasked questions by PLWHA and caregivers on the linkbetween nutrition and HIV/AIDS, dietary intake andfood access by PLWHA, ARVs and nutrition, tradi-tional herb therapy and maintaining body composi-tion. The counselling cards are designed for counsel-lors and health workers to help them in nutritioncounselling for PLWHA, such as making informedchoices for improving nutrition and using locallyavailable foods to meet their nutritional needs.

Copies of the materials can be obtained fromRegional Centre for Quality of Health Care(RCQHC), Makerere University Medical School, POBox 29140, Kampala, Uganda. Tel: 256-41-530888, Fax: 256-41-530876

• what side effects can be expected with ARV treatment

• interactions between ARVs and food, and how to cope with them

• how stigma can affect ARV treatment Further fact sheets are in production covering

other topics such as TB, prevention, monitoringtreatment, palliative care and living with achronic illness.

Participatory tools and activities, drawn fromAlliance workshops in several countries, will also bemade available later in 2005 to assist in educationabout ARV treatment and in developing ways ofsupporting people and communities with treat-ment.

The fact sheets are only available on-line and arenot in print. A series of 20 fact sheets are plannedand each one will be published on the Alliance web-site as it becomes available. There are currentlyseven fact sheets on the website, http://www.aid-salliance.org/sw19588.asp

For further information, contact: Garry Robson, Communications Assistant,International HIV/AIDS Alliance, QueensberryHouse, 104-106 Queens Road, Brighton BN1 3XF,United Kingdom. Tel: +44-1273-718-900 (main line), Fax: +44-1273-718-901 http://www.aidsalliance.org, http://www.aidsmap.com

News

33

This article is based on interviews with a number of indi-viduals within the Ministry of Health, in particular DrMichael O Carroll and Mrs Teresa Banda.

The HIV/AIDS pandemic has seriouslyimpacted negatively on all sectors of life inMalawi, irrespective of social class, and hasreduced adult life expectancy considerably.

With a population of around 10 million and aHIV/AIDS prevalence rate of 14%, it is estimatedthat around 180,000 people are at stages 3 or 4 in theprogression of the disease. As the country grappleswith the enormous strain of this pandemic, theMinistry of Health (MOH) has put together a strate-gy to deal with HIV/AIDS and its impact, with aconcerted campaign to scale up access to antiretro-viral drugs (ARVs) countrywide. Fifty-nine siteshave been identified within the country to supportARV distribution. It is planned to have 40,000 peo-ple on ARVs by the end of 2005 and a further 40,000by the end of the 2006.

Low CD4 count is one of the measurements fordeciding when to start patients on ARVs. Howeveras the equipment for measuring CD4 count is onlyavailable in six centres, the WHO algorithm (see box1) is being used for identifying symptomaticHIV/AIDS patients. The Malawi MOH has addedsevere wasting based on Body Mass Index(BMI<16) as one of the main criteria for diagnosis ofclinical HIV/AIDS, which is also indicative ofsevere malnutrition or severe wasting. Recent stud-ies have shown the benefits of nutrition support inconjunction with TB treatment, reducing mortalityand improving compliance. It has been recognisedthat around 70% of TB patients are HIV positive.

Nutrition StatusThe challenge to impact on nutritional status is

daunting, as many of the patients that start ARVtreatment are very ill and severely malnourished. Itis planned that these patients will be nutritionallyrehabilitated in a similar manner to the severelymalnourished children in the NRUs (nutrition reha-bilitation units), using F75 and F100 therapeuticmilks initially and later, using a Ready to UseTherapeutic Food (RUTF). It is not planned to reha-bilitate adult patients completely (as is currently thecase with malnourished children) but rather to sta-bilise their condition within a week to ten days andsend them home with RUTF and nutrition coun-selling.

The treatment of severe malnutrition in adults isnot as well researched and guidelines are less welldefined internationally, compared to managementof children. Consequently there are manyunknowns. Many severely wasted adult patientsmay not want to consume large volumes of milk. Itis also possible that a BMI of <16 is, in fact, too highand will lead to an excessive case load of malnour-ished people.

Staffing and health infrastructure capacityAlthough training has been conducted with

health personnel around ARVs, the training in asso-ciated nutritional care has not yet been carried out.The task is not easy, as human resource capacitywithin hospitals is severely stretched. Large hospi-tals requiring around 175 qualified trained nurseshave, in reality, a staff of only 25 nurses. During therecent food security emergency, it took a consider-able time to introduce the guidelines on treatment ofsevere malnutrition in children in the NRU’s, bothin terms of substantial initial training and thenongoing monitoring.

The attrition of health staff is due to many fac-tors, which include poor morale and working condi-tions, long working hours, low pay, lack ofresources, such as equipment, and lack of uniforms.There is also, like many other countries, a depletionof human resources due to emigration (brain drainto other African countries, Europe and the States)and the HIV/AIDS pandemic. It is generally recog-nised that the health infrastructure in Malawi is atnear collapse. A proposal has provisionally beenfunded by the Department for InternationalDevelopment (DFID-UK) to support reform of thehealth system, dependant on other funding from theglobal fund.

Bearing all this in mind, it is difficult to see how,with the best will in the world, the many challengesaround the health/nutrition care of PLWHA’S willbe managed.

PMTCT programmeThe PMTCT programme is an integrated

approach, working through the ante and postnatalclinics and offers a range of services. A large compo-nent of this is good nutrition education for allwomen attending the clinics. At present, not allfacilities have the capacity to do voluntary coun-selling and testing (VCT), but where these exist,women are encouraged to be tested - particularly ifsymptomatic. However, as it is voluntary testing,the wishes of patients are respected.

Women who test positive for HIV/AIDS receivefurther counselling on nutrition for themselves andon infant feeding. As 70-80% of Malawians live inrural areas with little access to clean water and elec-tricity and lack resources and purchasing power toobtain breastmilk substitutes, exclusive breast-feed-ing for six months is recommended as the safestoption. Furthermore, 90% of women choose tobreastfeed as it is the cultural norm and there is stig-ma attached to not breast-feeding. It is proving dif-ficult to get women to abruptly wean the infant offbreast-feeding at six months. One compromise maybe to support mothers to wean infants within a 2-3week period, once the child has reached six months.As complementary foods are renowned for theirpoor quality in Malawi, and in the absence of breast-milk, a study is being conducted into the use of aRUTF as part of the complementary diet of infantsover six months of age. Preliminary results suggestthat young children are growing very well with thistype of nutritional support.

In the three main hospitals in Malawi, WFP sup-ports the families of HIV positive pregnant women,with a family ration of 50 kg of maize, 7.5 kg of puls-es and 2.5 kg of vegetable oil. Mothers receive aration of 9kg of CSB (Corn Soya blend) and 900g900g of vegetable oil. It is believed that HIV positivepregnant women are particularly vulnerable, owingto the extra nutritional needs due to being HIV pos-itive, coupled with the additional nutritionalrequirements of pregnancy. The WFP ration contin-ues up to 18 months after delivery, to support thewell being of the child as well as the mother. Duringthis time, the infant is closely monitored, ensuringimmunisations (EPI) are completed. If deemed nec-essary, testing for HIV takes place. This programme

is a collaborative approach between the MOH andCHAM (Christian Health Association of Malawi) -the main implementers, UNICEF, WFP and NGO’ssuch as MSF-France and MSF-Luxembourg.Currently, WFP is supporting 3450 women and theirfamilies in this programme (Jan 2005).

As stigma remains a major issue in Malawi, itwas interesting to see how this was beingaddressed. As targeting certain individuals withfood would draw attention to these beneficiaries, insome clinics, women who have tested positive forHIV attend the clinic on a different day to the maincaseload of clients. This appeared to get round theproblem of stigma.

Home Based Care (HBC)General support to the chronically ill in the com-

munity falls under the umbrella of home based care(HBC). These initiatives are supported by a varietyof NGO’s, community based organisations (CBOs)and church organisations. The degree of linkage tohealth infrastructure varies significantly. In an idealHBC scenario, patients receive food (during crucialperiods), health care and psychosocial care as neces-sary, and support to improve longer-term house-hold food security through diversification andincome generating activities.

Food insecure households taking care of chroni-cally ill patients are targeted for support, as are peo-ple living with HIV/AIDS (PLWHAs) in recognitionof their higher nutritional requirements, particular-ly with regard to the need for adequate intake ofhigh quality protein and increased energy. The foodbasket is the same as that under the PMTCT pro-gramme and includes a household ration of 50kg ofmaize, 7.5kg of pulses, 2.5kg of vegetable oil, anindividual patient ration consisting of 9Kg of CSBand 900 grams vegetable oil.

In Malawi, WFP is working in partnership withNGOs in 11 districts targeting PLWHAs and chron-ically ill patients. The project is implementedthrough NGO’s, Community HBC volunteers andassociations of PLWHA, who already have a closelink with chronically ill patients and PLWHAs.

Regional resourcesDue to collaboration and co-operation with

other country programmes, resources that havealready been developed are being modified for thespecific context of Malawi. These include nutritionguidelines for PLWHAs and leaflets on topics suchas positive living, diseases such as diabetes, etc. In-country material is also being developed for use at‘grass roots’ level, such as ‘counselling cards’.

Conclusions and issuesMajor initiatives are underway to address some

of the issues around HIV/AIDS and nutrition sup-port. However as it is early in the process, there islittle evidence of impact of these interventions.Anecdotal evidence suggests that food aid/nutri-tion support, leading to longterm food security pro-grammes, have a positive impact. There is good evi-dence to suggest that supporting TB patients withfood aid reduces mortality and increases compli-ance. Many of the TB patients are also HIV positive(an estimated 70%).

In a country where food insecurity has been amajor issue for many years, addressing the nutri-tional needs of PLWHA is a considerable challenge,particularly as their nutritional needs are increaseddue to the illness. A recent study indicated thathouseholds with PLWHA’s and households withorphans and vulnerable children are more foodinsecure.

Internationally, nutrition in the context ofHIV/AIDS has been seriously neglected. This is alsoevident at a country level, where training aroundARV’s has already taken place but training in asso-ciated nutrition activities is planned at some futurepoint.

It is unclear if a BMI <16 is the most appropriatemarker for admission to hospital of suspected HIVpositive patients (based on clinical signs and symp-toms) for therapeutic nutrition support, as all theseextra admissions may completely overwhelm analready fragile health infrastructure. Importednutrition products, such as Plumpy’nut and thera-

WHO algorithm to identify HIV/AIDS

The WHO algorithm combines the pres-ence of two major signs and two minorsigns, if there is no other known cause ofimmuno-suppression. The symptoms are:

Major signs

• Weight loss or abnormally slow growth• Chronic diarrhoea > 1 month• Prolonged fever > 1 month Minor signs• Generalized lymph node enlargement• Oro-pharangeal candidiasis • Recurrent common infections• Persistent cough• Generalized rash • Confirmed HIV in mother

Support forPLWHAs inMalawiBy Mary Corbett

Mary Corbett is a food security and nutritionconsultant who visited the region on behalfof ENN in early 2005.

Views

34

peutic milk are extremely expensive commodities inthe context of an already extremely overstretchednational health budget. Is this the most appropriateuse of scarce resources?

Large volumes of milk as a stand-alone nutritionsupport may not be accepted by adults, and in somecountries, milk is considered as food for childrenonly. A combination of foods may need to be con-sidered, however there is limited research on thetreatment of adult malnutrition in these settings andit is unclear if there are budgets at hospital/cliniclevel to support nutrition through the use of suchfood commodities.

Malawi is probably one of the most advancedcountries with regard to strategic plans and guide-lines related to HIV/AIDS programming for differ-ent contexts. However, at a grass roots level where

This viewpoint is based on a series of meetings with keystakeholders in South Africa during Mary Corbett’s fieldtrip earlier this year.

As the HIV/Aids pandemic continues inthe Horn and Southern Africa, the preva-lence rates vary from place to place butthe impact can be felt across the region.

Although countries with small populations such asSwaziland and Lesotho are highlighted due to theirhigh prevalence rates, countries like South Africa,with somewhat lower prevalence rates but a farhigher population size, have far greater physicalnumbers of people living with HIV/AIDS – estimat-ed at 5.9 million people infected.

The situation in South Africa is complex. It is acountry with huge disparities of wealth. Althoughplaced in the upper half of UNDP’s Human Povertyand Development Index, over 50% of the popula-tion live below the poverty line. The country has atwo-tier health system with highly sophisticatedhospitals and clinics in the cities but poorlyresourced primary health care in remote rural areas.

Work on Nutrition and HIV/AIDSPrior to 1994, nutrition activities in South Africa

were fragmented. However since 1994, the Ministryof Health (MOH) has been involved in developingan integrated nutrition programme, whose mainstrategies include:

• Micro-nutrient Malnutrition Control• Food Service Management• Contribution to household food security• Disease specific nutrition support, treatment

and counselling• Growth monitoring and promotion• Nutrition promotion, education and

advocacy• Promotion, protection and support of

breastfeeding

The development and implementation of thestrategy has been a multi-sectoral approach involv-ing other ministries, such as such as the Ministry ofAgriculture and Ministry of Education.

In April 2004, the South Africa NationalGuidelines on Nutrition for people living withHIV/AIDS were completed. These guidelines areclear and easy to understand. The main emphasis ison maintaining a well balanced diet to maintainweight and keep healthy for as long as possible.Tips and suggestions are given on how to deal nutri-tionally with complications such as diarrhoea, nau-sea, vomiting, and mouth ulcers.

Although nutrition has been identified as a key ele-ment in the support to PLWHA, it remains difficultto address this area as there are many unansweredquestions including:

• What vitamin/mineral supplements to administer

• Which patients to give supplements to, and for how long

• Whether to give multivitamins as a blanket approach

• What is the impact of social services supplied food baskets and what should be included in the basket

While there are many mechanisms in place tosupport PLWHA, there remains awareness thatmore research is needed into areas around optimalnutrition requirements.

Disparities in needOn the one hand, there is great interest in food

supplementation and optimum multivitamin sup-plementation in South Africa. However, in a countrywhere half of the population is below the povertyline, household food security and well-balancednutritious meals are often not an option. Access tothe basics such as clean drinking water, health carefacilities and education remain out of reach formany. HIV/AIDS prevalence rates also vary consid-erably, in some ante-natal clinics, ranging from27.9% to 50% within the same area. The impact ofthese high rates on the community is tremendous,with vast populations faced with insufficient dailyfood intake.

The MOH in South Africa is striving to addressthe HIV/AIDS pandemic. Antiretroviral drugs(ARVs) are available free of charge in public hospi-tals and service points have been identified in eachdistrict to support ARV uptake. However, to date,the uptake has been surprisingly low. Although itwas planned to have 50,000 people on ARVs byApril 2005, by the beginning of 2005 this has onlyreached 15,000, despite an anticipated “rush to getfree ARVs”. It is unclear why uptake is so low in theface of substantial media coverage, however stigmaaround HIV/AIDS remains a major issue in SouthAfrica.

In spite of the considerable progress made toimprove health care and nutrition support overall inSouth Africa, HIV/AIDS has led to a substantialdeterioration in health indicators, such as infant andchild mortality, over the last number of years. Themost recent mortality figures (unpublished) indicatea serious increase, which is largely being attributedto the high incidence of HIV/AIDS. It looks like thechallenging, and perhaps unique, context of thecountry, with varying and substantial needs of vastsectors of the population, may well be limiting theimpact of mainstream developments in managingHIV/AIDS.

For further information, contact Mary Corbett,email: corbettmary@eircom.net

South Africa HIV/AIDS PandemicBy Mary Corbett

Mary Corbett is a food security and nutrition consultantwho visited the region on behalf of ENN in early 2005.

Dear Field Exchange,

In 2004 I attended a presentation onSprinkles, a micronutrient powder that canbe ‘sprinkled’ onto foods. The develop-ment of the product was supported by sev-eral large organizations in a not-for-profitmanner over the past several years. It was-n’t the first time I’d heard about the prod-uct as the group approached some organi-zations in Malawi a year or two before toexplore bringing the product here. Now, in2005, a message was posted to the ProNutHIV listserv - which links those concernedwith nutrition and HIV/AIDS - that theproduct aims to make it into all the homesof vulnerable populations.

I’m not at all opposed to sprinkles-types of products or other nutrient pills fortreatment (or in other special situations), aslong as recipients are also educated aboutwhere the nutrients come from in food. Butonce again, a message is being sent thatnutrition comes from a pill/a packet, a for-eigner, and all with money.

It is too bad that all that research, time,energy and money couldn't teach people(or local manufacturers) how to make theirown sprinkles from local nuts, fruits,greens, oilseeds, insects, fish, and the like.Instead of just sprinkling a packet onto abulky carbohydrate food, use the sprinkleas treatment along with teaching aboutplanting and eating less of that bulky car-bohydrate in the first place.

The results could be just as immediateand dramatic, but with an impact thatcould last for generations to come. Theorganisations that support this type of per-manent intervention could be mentionedduring every teaching session along withbig banners and flyers that announcedthem as the inventors and /or supporters.Just imagine a nice sprinkle powder thateveryone can have on hand to improvetheir own nutrition without relying on apacket from an outside source that is manu-factured with machines and jetted in withthousands of litres of petrol.

I'm sure that pre-packaged, sourced-from-far-away products have their place inwars, tsunamis, a few cities and other dis-asters, but for the majority of the 750 mil-lion children in the developing world, theirown indigenous foods would have just asmuch effect with a longer-term impact onthe society's nutritional health.

When I saw the developer of this prod-uct do a presentation on it, he did include asentence about diversifying diets as part ofthe whole project, but it was strongly over-shadowed by bringing in external resourcesand experts. When I asked him aboutusing the same resources that went intodeveloping, manufacturing and transport-ing Sprinkles to create a local ‘sprinkle’product with an emphasis on local diversi-fied diets, he immediately responded that itwouldn't work.

How do we know if no one really putsthe effort into it at the level that products,like Sprinkles, get?

Stacia Nordin, RD

Nutrition Consultant: Specialist in foodsecurity, sustainable agriculture,HIV/AIDSPO Box X-124, Post Dot Net Crossroads,Lilongwe, Malawi, email:nordin@eomw.net

Zlotkin S, Schauer C, Christofides A, SharieffW, Tondeur M,et al, (2005) Micronutrientsprinkles to control childhood anaemia. PloSMed 2 (1): e1. Source: PLoS MEDICINE, anopen access, freely available internationalmedical journal. http://www.plosmedicine.org

implementation occurs, there are significant con-straints, in particular around human resource capac-ity and availability of equipment and materials. Themajor concern is the capacity at ground level to sup-port this huge initiative, particularly as the healthinfrastructure is already completely overwhelmed.

For further information, contact Mary Corbett,email: corbettmary@eircom.net

1 Body Mass Index (BMI) calculated as weight (kg) divided by height (metres) squared

2 Plumpy’nut, produced by Nutriset3 See field article, Integrated PMTCT services in a rural setting

in Malawi, in this issue for experiences from St Gabriels Hospital, Malawi

4 See research article, Impact of HIV/AIDS on household food security and quality of life in Malawi, in this issue of Field Exchange

LettersViews

35

This article was written based on a WFP consul-tation to Malawi in February 2005.

The WFP Malawi HIV/AIDS project startedin November 2002 as a pilot project in fourdistricts, targeting 7,500 HIV/AIDS affectedhouseholds per year. During EMOP

(Emergency Operation) 10290, between July 2003and December 2004, the project expanded into 14districts to support 27,818 households caring for thechronically ill (CI) and orphans, and 10,000 individ-uals. In the current PRRO (Protracted Relief andRecovery Operation), the project will target 22,750households taking care of the chronically ill andthose keeping orphans and other vulnerable chil-dren, 1,250 TB patients, 3,450 HIV positive mothersenrolled in the Prevention of Mother to ChildTransmission of HIV (PMTCT) programmes, and5,300 chronically ill persons. Project support will bein 12 districts.

The overarching goals of the project have beento maintain the minimum acceptable nutritionaland dietary standards of PLWHA and other vulner-able groups through provision of information andservices on food aid and health. Specific objectivesare to:

• Improve the nutritional status of CI persons, including patients on Directly Observed TB Treatment (DOT) and PLWHAs on antiretroviral (ARV) treatment.

• Improve the short-term food security of households with CI persons and those taking care of orphans

• Strengthen the capacity of partner agency staff and community support groups to implement effectively comprehensive HIV/AIDS and nutrition related activities.

Programme targetingThe method of targeting for the programme was

elaborated in guidelines produced by a consortiumof INGOs, UN agencies and government ministries(JEFAP111).TheVulnerabilityAssessment Committee(VAC) report of February 2003 and NationalHIV/AIDS Surveillance report of 2001 were used toidentify the most vulnerable districts in the country,meaning those districts with:

• High numbers of cases of HIV infection (above 4% according to the 2001 National

HIV/AIDS Surveillance Report)• High prevalence of AIDS cases • High incidence of TB • High numbers of orphans • Food insecurity and low mean calorie intake per person per day• Existence of an NGO capable of

implementing HIV/AIDS activities and collaborating with other organisations in the district.

Priority Tribal Areas (TAs) were determined bythe lead NGOs through meetings with districtauthorities and other organisations working in thedistricts, to avoid duplication of efforts. Food inse-cure TAs were given priority, on the basis that thesecommunities have fewer resources to care for thechronically ill.

Beneficiary enrolment criteria at householdlevel included food insecure households that hadlimited or no source of income, and income withinthe lowest community category. Also identified bythe community or community organisation as inneed of food assistance were:

• Households caring for a chronically ill or bedridden patient (where chronically ill shall mean illness for one month or more).

• Households caring for a patient on DOT• Pregnant or lactating mothers under the

PMTCT programme and their babies• Households caring for orphans with special

attention to elderly, child or single headed households hosting a large number of orphans (as defined by the community).

The monthly food basket for those on theHIV/AIDS programme is 10kg of corn soyablend/vegetable oil for the chronically ill, and 50kgof maize, 5kg of pulses and 3.7kg of vegetable oil forhouseholds with CI or orphans.

Targeting took place through existing communi-ty structures, such as Village Action Committees(VAC), Orphan Day Care Centres or otherCommunity Based Organisations (CBOs) for thechronically ill or bedridden, and orphans. Targetingwas also advocated through institutions such ashospitals or clinics for TB patients on DOT andpregnant or lactating mothers under PMTCT andtheir babies.

Establishing the targeting process involved anumber of steps. Sensitisation meetings were heldat TA and village level to introduce the programme,discuss targeting criteria, and select or introduce anexisting committee. Orientation of committee mem-bers on their roles and responsibilities and identifi-cation and registration of beneficiary householdsfollowed. Verification of beneficiaries by NGOs incollaboration with the community or communityorganisations was conducted through communitymeetings or household visits.

Revision of guidelinesIn light of experience gained during EMOP

10290, including analysis of Post DistributionMonitoring and Community Household Surveill-ance data2 and an evaluation conducted in April/May 2004 (Salephera Consulting LTD 2004), theseguidelines were revised for the subsequent PRRO(JEFAP 111). Key findings in the ‘Salephera’ evalua-tion were that guidelines on targeting chronically illand orphans were followed in almost all areas visit-ed. However, difficulties implementing the guide-lines arose since there were more vulnerable peoplein the communities than the number actually cov-ered by the programme. Community perception ofthe most vulnerable for food aid has included peo-ple who are not most vulnerable as defined in theguidelines, and there has been limited communityinvolvement, sensitisation and verification duringbeneficiary targeting and selection. As a result, con-flicts sometimes developed between those responsi-ble for food distribution, beneficiaries, and thosethat believed they should qualify for food aid butwere excluded.

The new JEFAP 111 guidelines have, therefore,strengthened guidance on community sensitisation,selection of committee members, development ofcommunity defined selection criteria, beneficiaryselection and verification. It also made certain sub-stantive changes with regard to how to target.

Programme observationsA number of observations were made, based on

interviews with WFP staff and four site visits.

In the sites visited, those beneficiary chronicallyill and orphan containing households interviewed

WFP HIV/AIDSProgramming in MalawiBy Jeremy Shoham, ENN

1 Joint Emergency Food Aid Programme 2 See evaluation in this issue of Field Exchange

Households headedby the elderly may

be vulnerable

A N

ichols

on,

Mal

awi, 2

005

Field Article

were acutely food insecure and deserving of sup-port.

Although the programme for the CI and orphancontaining households should be integrated withother HIV services (see JEFAP 11 and 111 guidelines3),this is often not feasible. Hence, many beneficiariesof the CI and orphan programme receive food aidbut no other resource or form of support.Households and Village Relief Committee’s (VRCS)prioritised the need for drugs for these individualsto complement food provision.

The ration provided for households withorphans was the same, irrespective of the number oforphans. Households selected with orphans tendedto have multiple orphans, with an average of three.

None of the chronically ill or orphan containinghouseholds interviewed were aware of exit criteria,other than when the EMOP ended. It appears that,to date, there has been no graduation due toimproved food security of the chronically ill ororphan containing households.

Most beneficiaries (individuals or households)indicated that there were many others chronically illor households with orphans who were not enrolledon the programme but were equally needy. In acountry where an estimated 65% of the populationlive below the poverty line and there is an HIVprevalence of 14.4%, this is unsurprising.

Out of those presenting for selection for thePRRO starting in January 2005, the number/propor-tion of households per village selected wassmall/low (in some cases, each village was onlyable to select five households).

Ensuring that food commodities are deliveredclose to where beneficiaries live poses a consider-able challenge of loading small tonnages for variouslocations, which is not cost-effective.

Conclusions Donor environment and availability of resources

The current donor environment in the SouthernAfrican region is, at best, sceptical and, at worst,negative, with regard to the potential role of foodaid in supporting PLWHA – especially in non-emer-gency contexts. There are emerging views that suchhouseholds and individuals urgently require med-ical intervention and safety net programmes as apriority, rather than food aid. The result has beenthe type of situation seen in Malawi, where scarceresources from donors have to be targeted to a verysmall percentage of those in need, creating enor-mous pressures on WFP, implementing NGOs andcommunities.

However, food aid may well prove to be an inte-gral component of what PLWHA require. There isemerging evidence that certain nutrients may haltprogress of the disease, and adequate diet is essen-tial for maximising the impact of ARVs. Food aidmay assist compliance with treatment (ARV, DOT)and participation in programmes (such as PMTCT).Food aid may also be an important vehicle to allowacquisition of skills and community resourceswhich promote longer-term food security amongstPLWHA.

Yet, before such such programming can be fullysupported and rolled out on a national scale, thereneeds to be proof that objectives can be met. Longersurvival times, less morbidity, improved nutritionalstatus and attainment of longer-term food securityneed to be proven. Objectives of programming needto be clearly articulated and credible monitoringestablished to show whether these impacts areachievable and what the role of food is in meetingthese objectives. Perhaps key to proving a role forfood aid is the axiomatic truth that food aid can onlyhave a significant impact if well integrated withother services, including health care (for example,drugs for opportunistic infection and where possi-ble, rolling out of ARVs), health and nutrition edu-cation, psycho-social support, provision of adequatewater and sanitation, etc). The institutional com-plexities of providing integrated programming in a

country like Malawi is, however, poorly under-stood, both in terms of ‘how to make it happen’ andresources needed.

Rolling out a national programme before thereis evidence that objectives can be met within an inte-grated programming environment, and how torealise this environment, may set back a realisticappraisal of the potential role for food aid in sup-porting PLWHA.

TargetingIt is extremely difficult to evaluate the success of

targeting under the HIV/AIDS programme inMalawi and other countries in the region. While thePost Distribution Monitoring (PDM) examinesinclusion and exclusion with regard to social andeconomic criteria, it is unable to do this with regardto HIV infection as there is limited HIV testing inMalawi. While targeting through DOT, ARV,PMTCT and HBC programmes can be assumed toresult in high targeting efficiency with regard toHIV/AIDS infected individuals (over 70% of TBcases in Malawi are known to be HIV positive), tar-geting on the basis of chronic illness or havingorphans in a household may be an extremely impre-cise way of providing food security support target-ed at the most vulnerable PLWHA.

Although the JEFAP guidelines go some way toaddressing the above concerns in deriving andadvocating complementary indicators for inclusion,e.g. economic indicators, there is clearly significantroom for inclusion and exclusion error with respectto the chronically ill HIV affected and those house-holds whose food security has been critically under-mined by HIV/AIDS.

A recent study4 by AAH in Malawi drew the fol-lowing conclusions regarding targeting vulnerablehouseholds within the context of HIV/AIDS;

• Is it a greater priority to improve the selection of the more vulnerable households within Malawi’s context of chronic poverty, whether the vulnerability is rooted in HIV/AIDS or not? For methodologies intended to target food security input interventions, the degree of household vulnerability should take precedence over the cause of that vulnerability during beneficiary selection and should avoid reliance on simple proxies.

• Popular proxies used to detect vulnerable and/or HIV/AIDS infected/affected households, while theoretically indicative, are often practically imprecise with regard to identifying vulnerable households. Other proxies, such as the household food requirement:labour dependency ratio, should be piloted and developed.

RecommendationsTargeting food aid support to PMTC/ARV

/DOT individuals is an efficient means of providingnutritional support to the HIV infected. However,although consensus regarding the differential nutri-tional requirements and rations for those infected

with HIV has not been reached, there is some agree-ment on the need for differential rations accordingto stage of disease. For example, it is believed that a10% increase in energy intake is required to main-tain nutritional status and avoid weight loss ofasymptomatic individuals living with HIV, whilethose with AIDS related illnesses require a mini-mum of 20% increase in energy intake, with asmuch as 50% higher protein requirements. WFP willneed to monitor emerging consensus on this subjectand adjust rations accordingly. At the same time,implementing ration differentials will be extremelychallenging from a logistical perspective.

There is an urgent need to pilot integrated pro-gramming in one or two districts. These pilotsshould set clear objectives (nutritional, health andfood security) and establish rigorous monitoring.Such piloting was initiated in the WFP country pro-gramme before the emergency, but ceased with theadvent of the EMOP 10200 as other more pressingneeds took priority. WFP should also document theprocess and lessons learnt regarding how to estab-lish integrated programming and the costs andexpertise required to bring this about. If the findingsare positive, then donor organisations may be moresupportive of this type of programming with posi-tive implications for resources and future targeting.

If the decision is taken to continue with target-ing on the basis of CI and orphan containing house-holds, then this should only be implemented wherefully integrated programming can be guaranteed,i.e. it is combined with health service provision andeffective and proven food security support initia-tives. Future programming should, therefore, bebased on a mapping exercise to determine whereintegrated programming can be guaranteed. Thiswill lead to an overall smaller programme butshould also make it possible to target food resourceseffectively. Furthermore, committees at those sitesselected will not be required to make ‘difficult andpolitically sensitive’ household targeting decisions,as all CI and orphan containing households can beincluded. This type of ‘integrated and targeted’ pro-gramme would also substantially reduce logisticalcosts on a per tonnage basis.

Simultaneously, there is a need for pilot studiesto test the targeting efficacy of using proxies such asCI. Thus, a pilot study checking the serum status ofthe CI would be valuable, recognising that such astudy poses substantial ethical and practical diffi-culties.

For further information, contact JeremyShoham, ENN, email: jshoham@easynet.co.uk

3 For JEFAP guidelines, contact ALNAP, ALNAP Secretariat, ODI,111 Westminster Bridge Road, London SE1 7JD, UK Tel: + 44 (0)20 7922 0300, Fax: + 44 (0)20 7922 0399, Email: alnap@odi.org.uk or visit the weblink: http://www.alnap.org/pubs/pdfs/JEFAP_manual.pdf

4 Munk M and Fisher N (2004). Targeting vulnerable households within the context of HIV/AIDS. An evaluation of a piloted methodology. Ntchisi, Malawi. Action Against Hunger. See summary in this issue of Field Exchange.

36

Pregnant or lactating mothersmay be targetedwith food aid

Van

essa

Vic

k/W

FP

Field Article

In this article, EI shares their experiences of workingwith a community whose active involvement in reliefhas played a pivotal role in responding to chronic foodinsecurity complicated by HIV/AIDS.

Emmanuel International (EI) is a faith basedinternational NGO, with its head office inToronto, Canada. Emmanuel Internationalis operational in 11 countries worldwide,

with a major programme in Malawi. Other pro-gramme countries include Tanzania, Uganda,Sudan, Haiti, Brazil and the Philippines. Formed in1975 in response to several disasters around theworld, EI’s strategy has always been to workthrough local partners, strengthening the church to“meet the spiritual and physical needs of the poor inaccordance with the Bible.”

Work in MalawiEmmanuel International became operational in

Malawi in 1986, after a request from the EvangelicalBaptist Church of Malawi to help them respond tothe Mozambique refugee crisis when over a millionpeople crossed the border into Malawi to escape thewar. EI assisted the church with resources and staffto help meet the needs of the refugees in one area,and then gradually began development and spiritu-al projects, working with the local populationthrough the church.

After the war finished in 1992, the refugeesreturned back home and EI continued its develop-ment work in Malawi, while also becominginvolved in food relief work with WFP. The maindevelopment projects being implemented by EI arein the areas of food security, health and nutrition,water/sanitation and fertiliser for work. CurrentlyEI is working in four districts of southern Malawi,namely Zomba, Machinga, Mangochi and Balaka.The head office in Malawi is in Zomba town, the for-mer capital. Paul Jones, who is from Canada, is thecountry representative. Over the years, the projectshave grown or shrunk, depending on opportunityand the availability of funding. At present, theworkforce numbers around 160 national staff,assisted by eight international staff.

EI and relief workDuring the 2002/3 food security crisis, EI

became operational in relief once again, and wasinvolved in the formation of the Joint EmergencyFood Aid Programme (JEFAP) - a consortium ofNGOs in Malawi working together on WFPs EMOP(emergency operation) general distribution of foodaid to vulnerable populations. EI also becameinvolved as an implementing partner in the USAIDfunded C-Safe1 group of NGOs. While the large-scale general food distributions have been phasedout, EI remains an implementing partner with WFPunder their three-year PRRO (Protracted Relief andRecovery Operation) in Malawi. EI has also ongoingpartnerships with USAID and other NGOs in thenew five year Development Assistance Programme(DAP).

Challenges of implementationMalawi faces a complex mix of problems that

reinforce poverty and inhibit development. A highpopulation density - particularly in the south -together with depleted soils are, on their own, arecipe for chronic food insecurity. Other factors, likethe prevalence of HIV and lack of crop diversifica-tion, exacerbate the problem. Dependency on aid issomething all parties want to avoid, so all WFPspartners in Malawi are encouraged to get maximum’added-value’ for the food given, by forging link-ages between food-for-work (FFW) type projectsand other interventions. In one example from EI’sprojects, participants in FFW carried out the heavywork of cultivation on community gardens whileother households, targeted for aid under orphans orchronically ill projects, assisted with light work.Thus both benefited from the food produced. Thevillage chief has made land available to grow maize,cassava and rice, along with vegetables that areplanted in the dry season.

Involving the community in the targeting hasbeen essential to the acceptance of the programmes,particularly as resources can seldom meet the needsof every household that falls into the target catego-ry. With assistance from EI’s field staff, the commit-tee in each village consider the following criteria toselect households that receive assistance:

• Food insecurity, with a range of inclusion and exclusion criteria• Income-generating capacity of the household• Households caring for chronically ill persons• Households keeping orphans or other

vulnerable children.

In Mbalika village where EI are working, thoseinterviewed stated that there were around 30 chron-ically ill HH (households) and over 40 HH withorphans and vulnerable children (OVC) out of atotal of 378 HH in the village. However, they onlyhad resources to support 10 chronically ill and 17OVC households. The committee has had to makehard choices with regard to targeting the neediest,particularly as there is no home based care (HBC)system in the village, making decision making allthe more difficult.

Pilot programme to support TB treatment EI has also been part of a pilot project with WFP

and the national tuberculosis (TB) programme,assisting people diagnosed with TB in four selecteddistricts of Malawi. Patients diagnosed with TB atthe Machinga District Hospital are placed on theroutine 8-month directly-observed treatment (DOT)regime. Through EI, WFP provides a monthlyhousehold food ration to each household with a TBpatient, part of which is a nutritionally-dense cornsoya blend (CSB) plus a vegetable oil mix intendedsolely for the patient. EI receives food from WFP, re-packages it into the exact ration sizes, and deliversit to the local health centres around the district.Thus, when the patient goes to collect his month’ssupply of medicine and to be weighed, he or shealso collects a monthly food ration.

This intervention is being tested to see whetherit helps in two ways, first by encouraging patients tostick with the treatment regime until the course iscompleted, and secondly to see if the food reducesthe high death rate previously experienced in theearly months of treatment. Once the scheme hasbeen operational for a period of time, a direct com-parison can be made between a cohort of patientswho have received the food aid, and another non-recipient group that has never received food as partof the programme. Anecdotal evidence from healthcentre staff indicates, so far, that the food aid is hav-ing two effects. Patients diagnosed are beingretained in the treatment programme much betterthan before. Also, as news of the food entitlementhas spread, more people with symptoms like long-lasting coughs are presenting themselves for diag-nosis, in order to try and qualify for the food!

The collection and analysis of the data on deathsduring treatment is still being carried out and apaper on this should be published later this year.Even if, as expected, a reduced death rate during thetreatment period is found, it will require furtherinvestigation to discover exactly how the food hashelped. Is there a direct benefit to the patient interms of better nutritional status helping them totolerate the drug treatment? Or is it simply a case ofthe food acting as in incentive which keeps thepatient on the treatment? All that can be concluded,so far, is that both the health workers, and the 370 orso patients currently receiving food, are very happyto be involved in this project.

For further information, contact EmmanuelInternational Malawi Office: Private Bag 12, Zomba, Malawi, email: ei-malawi@malawi.net, website: www.e-i.org

1 Consortium for the Southern Africa Food Emergency

EmmanuelInternationalBy Andrew Mellen,Emmanuel International

Andrew Mellen has beenthe relief programmemanager with EI Malawisince 2003. With a background in agriculture,he previously worked asan organic farm managerin the UK. He currentlylives in Zomba, Malawiwith his wife and threechildren.

37

HBC in Mbalika villageM

Corb

ett,

EN

N,

2005

M C

orb

ett,

EN

N,

2005

Clients receiving HBCin Mbalika village

TB patient(BMI 13.9)

with hiswife

M C

orb

ett,

EN

N,

2005

Field Article

Malawi, a land-locked country in south-ern Africa that is prone to food short-ages, is bearing the full brunt of theAfrican HIV/AIDS pandemic, while

also struggling to strengthen its democratic institu-tions after 30 years under an extremely repressivedictatorial regime. In Malawi, where nearly 80 per-cent of households rely on agriculture as their majorlivelihood strategy, HIV/AIDS is having devastat-ing consequences. However, by building on theexisting capacity of local communities, CatholicRelief Services (CRS) programmes are creating pos-itive changes in the lives of vulnerable Malawians.

CRS in MalawiIn 1997, the Episcopal Conference of Malawi

(ECM) extended a formal invitation to CRS to workin Malawi. With this invitation, CRS first respondedto the needs of Malawians through small-scale ini-tiatives addressing food insecurity. Since then, over60 national and international staff members, locatedin the main office in Lilongwe and sub-office inBlantyre, have contributed to the expansion of pro-gramming. Developments include long-term sus-tainable interventions in agriculture, health andnutrition, HIV/AIDS, justice and peace. CRS’ pri-mary implementing partners in Malawi include theCatholic Development Commission of Malawi(CADECOM) and the Catholic Commission forJustice and Peace (CCJP).

Context of programme approachHIV/AIDS and related diseases are now the

leading cause of adult mortality in Malawi. TheNational AIDS Commission (NAC) estimates thatapproximately 15 percent of Malawians among themost productive age group (15-49 years) are infect-ed and, in urban areas, the level of infection inadults reaches 20 percent. Every year, as many as100,000 new infections occur, and at least half ofthose are among people aged 15-24 years. The three

greatest impacts of the HIV/AIDS pandemic inMalawi noted by NAC were:

• 800,000 of Malawian children under the age of 15 years have lost at least one parent to the disease

• the death rate has tripled since 1999 • the number of tuberculosis cases is three

times higher than it would be without AIDS.

Households affected by HIV/AIDS - whethercaring for and supporting orphans or a chronicallyill relative or neighbour - represent a staggering 64percent of the population in central Malawi. Loss ofproductive labour is the most direct and significantimpact of HIV/AIDS on these rural householdswho rely on agriculture. Extremely poor and vul-nerable households possess no excess capacity tosurvive additional burdens. HIV/AIDS is alsodiminishing the human capital of upcoming gener-ations as children, particularly young girls, aretaken out of school to care for sick family membersor sent out to work in order to subsidise familyincome. This severely limits their ability to gaineducation and life skills. Intergenerational knowl-edge is also deteriorating as parents die before pass-ing on wisdom and learning to their children.

Within this context, CRS/Malawi strives to sup-port individuals affected by and infected withHIV/AIDS. The agency will continue to support ini-tiatives that complement the goal of helping thepeople they serve to better cope with their situationthrough the following approaches:

• Improved quality and increased scope of community-based home care

• Increased support to orphaned and vulnerable children

• Empowerment for affected communities through integration of HIV/AIDS with other programmes

• Increased advocacy at all levels of society on addressing stigma and discrimination.

Dedza Integrated HIV/AIDS ProjectA significant component of the CRS operation in

Malawi is the Dedza Integrated HIV/AIDS Project.This has recently completed a three-year pro-gramme and intends to continue in the future withincreased activities. The project is implemented innine townships in Dedza, Ntcheu, and SalimaDistricts and strives to minimise the impact ofHIV/AIDS within the Dedza Diocese. Throughhome-based care volunteers, infected and affectedpersons receive care and support and communitiesreceive messages regarding HIV transmission.Additional components of the project includeincome-generating activities (IGA), distribution offood commodities to people living with HIV/AIDS,promotion of community-based HIV/AIDS educa-tion, and provision of vocational training for olderorphans. The project has indirectly touched 200,000of the 900,000 people living in the Dedza Dioceseand has 6,000 direct beneficiaries, includingorphans, chronically ill persons, widows, andHIV/AIDS infected families.

Home based careHome-based care (HBC) is the main component

of the Dedza Project. It is a community-basedapproach to providing health care and support tochronically ill persons and people living withHIV/AIDS (PLWHA). The programme operates inthree deaneries, with each deanery comprising oneHBC provider, one medically trained project officer,and 90 volunteers. The community identifies volun-teers for the programme. The volunteers are bothwomen and men and usually work in teams of two.Each volunteer supports three to five clients andmakes at least one visit per week to a client. Duringclient visits, the volunteer educates clients and othermembers in a household on primary health care,good nutrition, and HIV/AIDS prevention andtransmission. The volunteer also assesses the condi-tion of the client and provides basic medicines andother items to provide comfort. These items includepain relievers, malaria treatment, oral rehydrationsalts, antiseptic ointment, bandages, plastic sheets,gloves, and disinfectant. The volunteers will also dohousehold tasks if necessary (e.g. sweep, clean,cook, fetch water and or firewood) depending onthe condition of the patient.

HBC volunteers regularly gather for refreshertraining on palliative and home-based care. Atthese meetings they have the opportunity to discussany issues regarding their clients or workload.Traditionally, volunteers are permitted access tosmall-scale income generating activities in returnfor participating in the programme. In addition, thevolunteers maintain an elevated social status andare respected within their communities. As such, theretention rate for the volunteers surpasses 90 per-cent, which is quite high for a HBC programme.Lack of adequate food security and not having suf-ficient medicines are the main problems that themajority of volunteers report on behalf of theirclients.

Targeting beneficiaries Communities select the HBC beneficiaries for

the project. In general, the first layer of targeting forthe project is relatively easy, as the communitiesidentify households that have chronically ill house-hold members. The volunteers then visit the house-hold to determine if that household genuinely does

HIV/AIDS and FoodSecurity in Malawi

By Kathryn Lockwood, Martin DavidsonMtika and Richard Mmanga, CRS

Kathryn Lockwood is a nutritionist and is theHealth and Nutrition Programme Manager forCRS/Malawi.

Martin Davidson Mtika is a public health specialistand is CRS/Malawi’s Deputy Head of Programmingresponsible for Health, HIV/AIDS & Advocacy.

Richard Mmanga is CRS/Malawi’s Senior HIV/AIDSProject Officer.

38

A field of crops as part ofa food security project

M C

orb

ett,

EN

N,

2005

Rearing rabbits as part of foodsecurity interventions

M C

orb

ett,

EN

N,

2005

Field Article

have a chronically ill household member, and notjust an individual suffering from a short-term dis-ability. Communities readily accept this aspect oftargeting in the project. However, it has becomeapparent that communities were unclear as to thetargeting criteria surrounding the Orphan andVulnerable Children (OVC) component of the proj-ect, which led to a low-level tension in the commu-nities between beneficiary OVC households andnon-beneficiary households. Given this, in thefuture the project will hold community educationand sensitisation meetings with each of the targetedcommunities to clarify targeting criteria.

Obviously, some clients did not survive the fulllength of the project, and additional new clientsneeded assistance. Therefore, the same HBC clientsdid not remain in the project the entire time.However, community HBC volunteers largelyremained the same, and the same communities weretargeted throughout the life of the project.

Food aid componentLocal food, mainly maize, pulses, and fortified

grains, is regularly made available to extremely vul-nerable households. HBC volunteers demonstratemeal preparation to households using these fooditems along with other locally produced foods. Soymilk and porridge preparation is a new activity forthe project. Using approximately two kilograms ofsoy and nearly 2 litres of milk, enough porridge tofeed a large household can be produced. Volunteersalso prepare traditional meals with maize and showhouseholds how to increase nutrient values withlocally grown items. Community gardens also con-tribute to the food supply for vulnerable house-holds.

Income generating activities (IGAs)The project also supports income-generating

activities for older orphans and vulnerable childrenand households supporting chronically ill persons.Orphans and vulnerable children learn vocationalskills such as sewing, tin smithing, and carpentry.Upon completion of coursework, they receive mate-rials to get them started, such as basic tools andsewing machines. Household income-generatingactivities include rabbit breeding and agriculturalproduction . The rabbits mainly serve as food and,as they reproduce, are distributed to other house-holds, however they can be used as assets if needed.Agriculture extension workers introduce time- andcost-saving mechanisms to help increase crop pro-duction. Beneficiaries learn how to use manure ortheir own compost for fertilizing, crop rotation tech-niques, multi-cropping (growing different cropstogether in one field), and how to grow kitchen gar-dens. In some areas they also receive treadle pumpsfor irrigation. Based on the experiences beneficiaries

have had selling vegetables in markets, they havedecided to form cooperatives to increase their prof-its by reducing competition and working togetherto set prices. A new activity due to start this year ishoney production, which can be sold in the marketand consumed by chronically ill persons as an alter-native to refined sugar.

Impact of programmeThe end of project evaluation revealed that 43

percent of households in the target area reportedbenefiting from this project, an increase over theproject’s objective to provide care and support to atleast 25 percent of people infected and affected byHIV/AIDS. The project also aimed to provideHIV/AIDS education in the targeted area.Household surveys during the project evaluationdemonstrated that approximately 67 percent of thecommunities had benefited from education and pre-vention messages.

During the same evaluation, 95% of respondentsreported that they required additional food securityinterventions. Specific needs articulated by PLWHAincluded increased agricultural inputs, additionalincome generating activities, augmented supportfor orphans, more access to medicines, and scaled-up psychosocial support.

Interviews conducted by ENN with HBC staffin the field supported many of the CRS findings. Forexample, the main challenges identified by HBCvolunteers included;

• Lack of food and food security for clients• Lack of medicines (although volunteers have

been taught about use of local herbs for treatment of illnesses associated withHIV/AIDS

• Lack of clothes• Patient isolation• Lack of transport

An interviewed HBC co-ordinator describedelements of the programme considered to haveworked well, in particular patient care, includingmedication, orphan care, skills training, educationalsupport, community empowerment, and incomegeneration.

Youth mobilisation remains a challenge, whileforging linkages with health infrastructure althoughstarted late, has made some progress. Since a work-shop with staff from all the health institutions inoperational areas to sensitise them to the workbeing carried out by the HBC teams, some of thehealth centres now have an official referral letter sothat patients can access health care. However it wasfelt that these linkages need to be strengthened fur-ther.

Future plansCRS/Malawi plans to continue supporting this

project in the future and is currently adjusting theiractivity plan in conjunction with CADECOM basedon the evaluation. Namely, CRS and CADECOM arein the process of developing a more integrated foodsecurity and HIV/AIDS project. In addition, theorganisations are slightly modifying and scaling-upother interventions based on the results of the end-of-project evaluation.

One of the interventions the new project willscale-up is the OVC component of the project,which focuses on mitigating the impact ofHIV/AIDS on OVC and their quality of life. CRSand CADECOM will integrate food security, protec-tion, and psychosocial activities into the OVC com-ponents. In addition, CRS and CADECOM plan towork closely with the Malawian Ministry of Healthto ensure collaboration with the Government’s anti-retroviral therapy (ART) rollout for current HBCclients.

For further information, contact: Kathryn Lockwood, CRS/Malawi, Private BagB319, Lilongwe 3, Malawi. Tel: +265 1 755 534,email: klockwood@crsmalawi.org

In common with other countries in southernAfrica, the Malawian population is heavilyaffected by the HIV/AIDS epidemic. Themost recent sentinel surveillance report1

revealed a national HIV prevalence level of19.8% among antenatal care (ANC) attendants.Projecting these findings, the prevalence levelfor HIV infection in the adult population (15-49years) is estimated at 14.4% (12-17%)2. Approx-imately 110,000 new HIV infections occur annu-ally. In 2003, about 900,000 people were estimat-ed to be living with HIV/AIDS in the country, ofwhom close to 170,000 had advancedHIV/AIDS disease and were in need of life-sav-ing antiretroviral treatment (ART). Of these,10% are children, the majority having beeninfected through mother to child transmission.With an estimated 80,000 adult and childHIV/AIDS related deaths annually, HIV/AIDShas become the most significant cause of deathamongst adults - average life expectancy inMalawi has now dropped to below 40 years.Moreover, Malawi has, at present, approximate-ly 840,000 orphans3, 45% of whom have lost oneor both of their parents because of AIDS.

Fuelled by this epidemic, the annual casenotification of tuberculosis has increased from5,334 new cases (82/100,000) in 1985, to nearly30,000 new cases (257/100,000) in 2003. This iscoupled with a significant increase in mortality,to more than 20 %, among tuberculosis (TB)patients under TB treatment. The situation isfurther complicated by food shortages, whichMalawi is increasingly confronted with duringthe yearly hunger gap periods. It is feared thatthis precarious food security situation will leadeventually to a nutritional crisis situation, withhigh levels of acute malnutrition during theannual ‘lean’ season (rainy season) before theharvest.

MSF in MalawiMSF has worked in Malawi for many years,

first in the refugee camps during the Mozam-bican war and later, since 1997, focusing on the

Nutritional SupportThrough HBCin MalawiBy Mieke Moens, MSF

The author would like to acknowledgethe contributions of Dr. Roger Teck,Head of mission, Thyolo, Malawi andPascale Delchevalerie, nutritionist atMSF-Belgium headquarters.

Mieke Moens is apaediatric nurse,and is responsiblefor the PMTCT andnutrition pro-gramme for MSFThyolo, Malawi

39

1 At time of writing, the most recent report was HIV Sentinel Surveillance Report 2003. Malawi Ministry of Health & Population and National AIDS Commission, November 2003.

2 HIV and AIDS in Malawi, 2003. Estimates and Implications. National AIDS Commission. January 2004.

3 In Malawi, orphans are defined as children who have lostone or both parents (because of death) and who are stillunder 18 years old.M

Corb

ett,

EN

N,

2005

Women makingsoya milk

Field Article

HIV/AIDS pandemic. The current MSF programmeis based in Thyolo district, Malawi and comprises anumber of components including:

• Prevention of HIV transmission• Support to the district tuberculosis control

programme• Home based care (HBC) and social support for

people living with HIV/AIDS and/or TB.• Hospital based care for patients for HIV/AIDS

and/or TB• Integrated antiretroviral treatment programme

since April 2003• Emergency preparedness and response to

cholera outbreaks for the five most southern districts in the southern region of Malawi.

• Nutrition programme, whose key activities include:

- Running a Nutrition Rehabilitation Unit (NRU) in Thyolo District Hospital

- Supervising the Supplementary Feeding Centres (SFC) of Thyolo District

- Orphan’s programme for infants up to 1 yearof age

- Integrated nutrition support to malnourishedpatients in the HBC network

- Nutrition support to TB patients during the first month of treatment

- Nutrition support to malnourished patients registered with the ‘Continuum Care’ (HIV/ARV) Clinic

- Nutrition supplementation of malnourished patients hospitalised in Thyolo District Hospital

• Operational research to document and adjust approaches, to prove feasibility and to provide evidence for dissemination and advocacy at national and international level.

The HIV/AIDS related targets (2003-2007) forThyolo district are, access to a ‘continuum of care’for at least half of the estimated 50,000 people livingwith HIV/AIDS, and access to ART for at least 50%of the estimated 7,000 – 8,000 people living withAIDS. The target groups are malnourished peoplewith HIV/AIDS and/or TB, and people with a poormedical condition.

Integrating HBC and nutritionMSF currently coordinates and integrates HBC

and nutrition programmes in the district. The objec-tives of the programme are to reduce malnutritionamongst the chronically ill and to support malnour-ished AIDS patients during initiation of ART. MSFcurrently have 550 patients registered through HBCactivities and 33 through the antiretroviral (ARV)clinic (Jan 2005).

Integrated activities have been taking placemonthly over five days – a period that has recentlybeen extended to 10 days a month. Each month, thenutrition team joins the HBC team and the patientsare screened in their community (or the nearestcommunity where this activity takes place), whilethey are waiting for treatment from the HBC nurse.Screening is carried out by trained HBC volunteers.The results (height, weight and MUAC) are writtenup in their individual health passport. The nutri-tional team then evaluates each individual by calcu-lating the Body Mass Index (BMI), checking foroedema and evaluating the general health condi-tion. Where individuals meet the entry criteria, thepatient is admitted onto the programme (70% areadmitted on the basis of BMI<17). Each patient

receives a monthly ration of 10kg of Likuni Phala(Malawian fortified blended food).

Some of the ARV patients are not supportedthrough HBC but by the nutrition team in ThyoloDistrict Hospital. The same entry criteria apply forhospital nutrition support but patients are providedwith Plumpy’nut (Nutriset) instead of Likuni Phala.

Initially discharge criteria were established as aweight gain of more than 10% during two consecu-tive visits4 , a BMI >17 (or MUAC > 185) and goodgeneral health condition. However these provedimpractical, both in terms of calculating the 10%weight gain and the degree of weight gain, whichsometimes took too long to reach. The criteria weresubsequently revised to achieving a BMI >18.5 fortwo consecutive visits and a good general healthcondition. Generally, patients remain for a consider-able time on the programme (average stay in 2004was 5.5 months). Table 1 shows progress of patientsin a one year period.

ARVs for childrenMSF has recently started an ARV clinic for chil-

dren. It has been a challenge to establish appropri-ate entry criteria for children and nutrition support,especially for those aged between 7 and 18 years.Since MUAC is not an accurate measure for thisgroup, weight-for-height or BMI and the generalhealth status is used. For smaller children, existingentry criteria for NRUs and SFCs are applied. Thesecriteria are open to change, if experience leads us tobelieve that they are not appropriate.

The MSF team strongly believe that nutritionalsupport for this vulnerable group is necessary,although it is difficult to measure the impact.Implementation of a double blind control studywould be ethically difficult. The programme is alsoshifting slowly to using Plumpy’nut instead ofLikuni Phala, as it is a more nutritionally completesupplement, more manageable in terms of carriageand storage, and is easier for beneficiaries to use.

This programme has many challenges. Forexample, there is currently a low attendance rate insome areas, especially when the weather is poor orwhen there is a planting or harvesting period.People who send relatives, instead of attendingthemselves, for three consecutive visits are now vis-ited by the team or through HBC workers, to estab-lish the reason.

It is important that the objectives of this type ofnutritional support are realistically defined in rela-tion to access to medical care and treatment. InThyolo district, people often live far away from thehealth structures, so that transport is difficult.Consequently, monthly attendance is not alwaysrealistic. On the positive side, nutritional screeningcan contribute to the identification of HIV positivepatients eligible for ART (a wasting syndromeoccurs at stage 4 in AIDS). In conclusion, throughthis programme experience, we have demonstratedthat integration of well targeted nutritional supportfor malnourished people with HIV/AIDS is feasiblein a setting with community home based care.

For further information, contact: Mieke Moens,PMTCT and Nutrition programme, MSF Thyolo,email: MSFL-Blantyre@Luxembourg.msf.org

4 Visits take place once a month.

Table 1 Weight progress of those patients registered in nutrition programme

Total number of patientsevaluated

Number receiving ART Number not receiving ART

n 3294 325 2969

Weight gain

Weight static 29.5%

Weight loss 29%

41.5 % 56%

24.5% 30.5%

22% 29.5%

40%

Period: February 2004 – January 2005

40

Height beingmeasured bytrained HBCvolunteers

M M

oen

s/M

SF,

Mal

awi, 2

005

Team writing up individual'health passports'

M M

oen

s/M

SF,

Mal

awi, 2

005

Distributingthe monthlyration ofLikuni Phala M

Moen

s/M

SF,

Mal

awi, 2

005

Field Article

ing orphans, are not always the poorest and most food insecure.In particular, a significant number of orphans live in better offhouseholds. This poses a targeting problem since WFP is commit-ted to feeding only the hungry poor, and lacks the resources tocover all HIV/AIDS affected households in Busia. Thus, WFP hasapplied a dual targeting criterion, to identify households that areboth affected by HIV/AIDS and are food insecure. This is chal-lenging because targeting HIV/AIDS-affected householdS is verydelicate, with a risk of creating stigma. In addition, it is costly anddifficult to evaluate food security status at a household level.

ImpactAn impact survey was conducted in 2004 in five operational

areas. On both criteria, Butula Division had the highest target effi-ciency, i.e. the majority of the people receiving food assistancewere clearly both HIV/AIDS affected and food insecure in theabsence of food assistance. Only two percent of eligible peoplewere excluded from the programme. The partner in Butula is theRural Education and Economic Enhancement Programme(REEP), a CBO under the dynamic and inspired leadership ofMary Makokha. So what are the main ingredients of REEP’s suc-cess?

1. Community-driven and community based:The food is provided through elected Community ManagementCommittees (CMC), which are independent of the local politicaland patronage structures, and generally comprise affected orinfected people. The staunch independence of REEP ensures that,for the most part, food allocation is decided based on needs, andnot on influence. REEP officers know the community intimately,well enough to be able to verify the CMC participant lists.

2. Women in the driving seat: As with all WFP programmes, women’s participation wasencouraged and facilitated in all the stages of the Busia project.Over two-thirds of the CMC members are women. Women tendto have better information on the relevant household characteris-tics, and most community members consider the CMC to bemuch fairer than the male-dominated local leadership structures.

3. Tackling stigma:REEP has done an extraordinary job, building people’s self-esteem and enabling them to speak freely about their HIV status.When people are found to be HIV positive, they are referred to asupport group headed by a community health worker, and theirfood security situation is assessed in order to determine their eli-gibility for food assistance. This improves the quality of the tar-geting, by encouraging more destitute and infected people tocome forward.

With the right quantities of food getting to the right people,the impact of the food assistance has been very positive. The pro-portion of poor households consuming adequate kilocalories hasdoubled. Participants report a tremendous improvement in theirhealth, self-confidence, and entrepreneurship.

According to Mary Makokha there have been “resurrections”among the bed-ridden patients receiving CSB rations from WFP.On visiting Butula with ENN, WFP met a woman who was bed-ridden before the project but is now farming her own land, andhas even increased her plot size. Although the programme targetsthe poorest households in Butula, there were no children suffer-ing from wasting among those surveyed.

While the REEP experience has encouraged WFP to expandthe food assistance programme, the agency is aware of the needto complement short-term food assistance with longer-term foodsecurity initiatives. In Butula and other project sites, WFP part-ners are setting up income-generating projects, like poultry andpork raising. These activities do not require heavy labour, so theyare appropriate for households affected by HIV/AIDS.

Of course, effective targeting is not enough. In Butula, foodsupport is an essential tool for mitigating the food security impactof HIV/AIDS, but is only one input in the support package.REEP’s comprehensive home-based care programme offers tar-geted households a support package, which is complementedwith WFPs food assistance.

See related field article, REEP experiences in Western Kenya, p26

For further information, contact WFP Kenya, UN GigiriCompound, Nairobi, Kenya Tel: +254 20 622043/622233, Fax: +254 2 622263, email:WFP.Nairobi@wfp.orgThe authors can be contacted by email: KGarnier@faonairo-bi.or.ke, Ruth.Situma@wfp.org, Ben.Watkins@wfp.org

41

Participatory AFood Security

By Erin Culbertson andMoses Kalyebara

Erin Culbertson has been theTechnical Writer for Plan Ugandasince July 2003. As part of herprimary degree in Public andInternational Affairs, sheresearched community-basedmalnutrition initiatives in develop-ing countries and volunteered ata malnutrition centre in CapeTown.

Moses Kalyebara, an agriculturegraduate of Makerere University,Uganda, has worked for Plan for7 years as the CountryAgricultural Advisor. Before this,he was one of the managers of aBritish tea company and themanager of agro-processing foran INGO called World Learning.

The authors would like toacknowledge the contributions ofthe following to the project inUganda: David Kyeyune, ProjectCoordinator for Food Security,Cassiano Kansiime, CommunityDevelopment Coordinator (CDC),Martin Nzabala, (CDC), and thecommunities of Kasana andBamunanika in Luwero District.

SUDANK

ENYA

ZAIRE

UGANDA

WFP supports a project, Nutrition andCare of the People Affected byHIV/AIDS, implemented by theGovernment of Kenya, NGOs and

community based organisations (CBOs), in Busiadistrict in Western Kenya. Busia district was select-ed due to a high HIV sero-prevalence (over 15 per-cent) and relatively high rates of rural poverty. Theproject commenced in 2003 and provides a basicfood ration to 24,700 food insecure and HIV/AIDS-affected people. The ration meets 57 percent of thetotal dietary energy requirements of the targetedhouseholds. WFP’s food assistance complementsother HIV/AIDS mitigating activities in Busia dis-trict, including provision of anti-retroviral therapy.

Delicacy of targetingA baseline survey, undertaken prior to imple-

mentation, found that the households that are mostdirectly affected by HIV/AIDS, such as those withbedridden adults, with grandparent heads, or host-

ThreeIngredients of Success:Targeting FoodAssistance inWestern KenyaBy Karine Garnier, Ruth Situmu and BenWatkins, WFP Kenya

Karine Garnier (MPA) is a consultant now work-ing with FAO

Ruth Situma (MSc Public Health and Nutrition) isa nutritionist with WFP Kenya

Ben Watkins (MSc Ag. Econ.) is programmeadvisor for WFP Kenya

TANZANIA

RWANDA Plan Areas

Many handssharing the

family pot

Van

essa

Vic

k /

WFP

Vanillapods growing as part ofproject tosupportcash crops

M C

orb

ett,

EN

N,

2005

Field Article

Adap

ted f

rom

w

ww

.inte

rnat

ional

.egm

ont_

hs.

dk/

nudip

u/

42

Plan is an international, non-governmental,child-centred development organisationwithout religious or political affiliation (seebox for the basis of the organisation’s oper-

ations). For more than 12 years, Plan Uganda hasworked in partnership with Ugandan communitiesto establish successful health, education, incomegeneration and housing/sanitation improvementprogrammes that are meeting the needs of over300,000 people. This commitment to community-driven development has helped the organisationwin the trust and confidence of community part-ners.

Plan works in four districts in Uganda(Kampala, Luwero, Kamuli and Tororo), primarilylocated in the central and eastern regions of thecountry. Over 80% of families in Uganda rely onsubsistence agriculture. Low agricultural productiv-ity, degradation of natural resources, and limitedaccess to modern agricultural technologies and mar-kets are all factors that may reduce families’ house-hold incomes. It is difficult for families to raiseenough food for survival on this land, let aloneadditional food to create an income basis. Childrensuffer the effects of this poverty, as food insecurityleads to deficient diets and malnutrition.

Vision and basis of Plan’s workFundamental to Plan’s operations is the organi-

sation’s vision of a world in which all childrenrealise their full potential in societies that respectpeople’s rights and dignity

Child sponsorship is the basic funding compo-nent of the organisation, with over one million chil-dren and families enrolled. Plan strives to achievelasting improvements in the quality of life ofdeprived children in developing countries througha process that unites people across cultures andadds meaning and value to their lives by:

• Enabling deprived children, their families and their communities to meet their basic needs and to increase their ability to participate in and benefit from their societies

• Fostering relationships to increase understanding and unity among peoples of different cultures and countries

• Promoting the rights and interests of the world’s children.

Child Centred Community Development(CCCD) is Plan’s basis for planning, resource mobil-isation, implementation, and monitoring and evalu-ation of all programmes. It is equity-based andrequires a change in relationships between andamong individuals and institutions at all levels –children and adults, the poor and the elite, femalesand males. CCCD emphasises inclusiveness,

respect, shared learning and the importance ofactively engaging children and adults in mattersthat affect them. Plan has begun (and will continue)to shift its programme approach away from onebased upon traditional top-down delivery of servic-es, towards one that is more child-centred, partici-patory, and community-based.

The HIV/AIDS pandemic is one of the biggestobstacles to reaching poverty reduction targets anddevelopment goals in Uganda. The pandemic hasresulted in the death and illness of men and womenin their prime ages, thus reducing the productivelabour force that is engaged in agricultural produc-tion and also creating a large number of orphans inUganda. Some farming families affected by illhealth and death from HIV/AIDS have stoppedplanting traditional food crops, such as nutrient-rich beans, and have replaced them with less nutri-tious root crops that are easier to produce. When theprimary breadwinner of a family falls ill, the entirefamily’s food security is threatened. Additionally, aperson living with HIV/AIDS needs to maintain abalanced, adequate diet to boost her/his immunesystem. To make ends meet, many families are sell-ing off livestock, crops their children should be con-suming, and household assets. In order to addressthe impact of HIV/AIDS on agriculture, food secu-rity, and nutrition, crops and production technolo-gies that require less labour and inputs, yet retainmicronutrients, need to be promoted. Higher valuefood crops, such as coffee and vanilla, also will bepromoted to improve nutrition and income-generat-ing activities.

Food security project In response to these factors, Plan developed the

Strengthening Food Security for Children andFamilies Living in Poverty Project. The project isbeing implemented in one district, Luwero, which islocated about 45 km from the country’s capital. Itbegan in January 2004 and is projected to end inDecember 2008. The project was created throughfunding by the Douwe Egberts Foundation.

The overall project objectives include:

• To enhance awareness of coffee as a cash crop and improve coffee quality through training and support to small coffee farmers, utilising technologies for pre- and post-harvesting, with particular emphasis on female farmers.

• To augment market efficiency and farmers’ incomes through development and empower-ment of small groups of coffee farmers.

• To create and retain community participation through enabling families to make enhanced-productivity agricultural investments, to sustain efforts for environmental conservation, and to properly manage natural resources.

• To facilitate uptake of thorough farm manage-ment practices through the introduction of intensive farming techniques and income-generating activities.

• To maintain an outreach programme in schoolsto enable school children to improve their knowledge and skills in agriculture, environ-mental conservation and natural resources management.

• To promote childhood care and development inthe context of the communities’ needs and resources, with the ultimate goal of enhanced food security and improved nutritional status.

In Plan Uganda’s new Country Strategic Plan,the Sustainable Livelihoods Approach (SLA) is usedto understand and analyse the circumstances of vul-nerable children and their families and to identifythe families that qualify as ‘chronically poor’ andthose that are ‘economically vulnerable.’ Targetedlivelihood interventions for the families will be usedto improve the food security and householdincomes only for those families identified as ‘chron-ically poor.’ For the ‘economically vulnerable’ fami-lies, capacity-building exercises, agricultural train-ing programmes, and income-generating activitieswill enhance their abilities to provide for their chil-dren’s basic needs.

The ‘chronically poor’ include the severely dis-abled, terminally ill, child-headed households, theunemployed and the landless. People living withHIV/AIDS and their families will also fall into thiscategory, particularly as the parents become weakerand eventually die, leaving widows, orphans, andchild-headed households in precarious situations.This category of people has difficulty participatingin community-managed projects. Studies haveshown that targeted transfers of agricultural inputscan increase the livelihoods of the poor, particularlyif they utilise the transfers for investment andimproved productivity.

The ‘economically vulnerable’ include cash cropfarmers, orphans, informal sector workers, the eld-erly and widows. This group could potentially ben-efit from different interventions, such as micro-cred-it programmes, training in modern farming tech-niques, and capacity enhancing activities.

Community participationAfter raising awareness in communities about

the project, over 1,000 farm families were identifiedby their communities through the SLA. Using a par-ticipatory approach, a ‘social map’ was createdwhich included the number of people living inhouses, their livelihoods, and the heads of the fami-ly. Both the ‘chronically poor’ and ‘economicallyvulnerable’ were identified, as various project activ-ities could benefit the different groups. These partic-ipants received training on project structure, man-agement, documentation and objectives.

One quarter (25%) of the participants selectedwere women. This number is lower than targeted,because coffee is traditionally seen as a man’s cropand most women do not have ownership of the cof-fee gardens. Gender-disaggregated data was collect-ed on women’s involvement in and control of coffeefarming. The role of women in attaining food andnutrition security is extremely important, particu-larly because experience shows that women transferimproved food/nutrition security and income totheir children.

The project was developed and implemented incollaboration with local government structures, anNGO called Joint Energy and Environment Projects(JEEP), and Ibero (U) Ltd. In this project, Plan isadopting a community-managed project approachfor certain components, such as animal provision.The project has embraced the concept of farmer-to-farmer extension, in which model farmers voluntar-ily support approximately five other farmers.

Approach to in Uganda

Schoolboys registered in school feeding programme

M C

orb

ett

EN

N,

2005

Coffee plantationsupport group

M C

orb

ett,

EN

N,

2005

Field Article

43

This article describes the evolution of GOAL HIVrelated programming activities in Malawi over thepast three years, but does not describe the implemen-tation of or lessons learnt from these programmes(Eds).

GOAL and JEFAP

GOAL became operational in Malawi inApril 2002, and was involved in the for-mation of the Joint Emergency Food AidProgramme (JEFAP), a consortium of

agencies, including WFP and 11 other NGOs,formed to manage the distribution of food on anational basis. Operational in two districts,Blantyre and Chiradzulu districts, GOAL distrib-uted 2,300 metric tonnes (MTs) of food per month atits peak, before completing this programme phasein May 2003. Continuing with the consortiummodel for the second year under JEFAP 2, GOALwas engaged in food for work (FFW) and the directtransfer of food aid to People Living withHIV/AIDS (PLWHAs) and the chronically ill. Thisinvolved supporting over 9,700 households affect-ed by HIV/AIDS, with over 750 FFW projects oper-ating in Blantyre and Chiradzulu Districts ofMalawi. This provided the basis for activities thatwere implemented under JEFAP 3 in Nsanje dis-trict, where GOAL relocated (funded by WFP) tooperate in an area of acute need in the southern-most area of the country. FFW projects were imple-mented in all nine Traditional Authorities (TAs),with a range of activities designed to support com-munities in the medium and longterm. Theseincluded reforestation and community woodlots,rehabilitation of key roads, rainwater harvestingand soil and water conservation, along with othercommunity based agriculture projects. Food tothose affected by chronic illness (2,041 families) andto those looking after the chronically ill (360 fami-lies) was supplied in seven of the TA’s. This pro-gramme finished at the end of December 2004,coinciding with the beginning of WFP’s ProtractedRelief and Recovery Operation (PRRO) in January

2005.Goal and PRRO

The WFP PRRO is scheduled to run for threeyears, with the initial three month programme run-ning to the end of March 2005. During this period,4,807 MT of food was allocated to GOAL for distri-bution through FFW (26,110 households), food tothose affected by chronic illness (2,048 households)and those hosting patients (620 households). A totalof around 160,000 will benefit in this blanketresponse to what is traditionally a period of acutehunger as people wait to harvest their crops inApril. Additional funding from the Jesuits has beenreceived to support food to the chronically ill in thetwo TAs not served by the PRRO, up to the end ofFebruary 2005.

Complementary activitiesGOAL views these activities with WFP as the

stepping stone to consolidating a variety of activi-ties in Nsanje, with particular emphasis on foodsecurity, agriculture, drought and flood mitigation.

Fortified foodsDuring JEFAP 1, with funding from

Development Cooperation Ireland (DCI) and in par-allel to the general food distribution, GOAL distrib-uted Corn Soya Blend (CSB) to the most vulnerablegroups during the hungry season months ofJanuary to May 2003. Further DCI funding sawGOAL establish a supplementary feeding pro-gramme in 21 health centres and 63 outreach clinicsin Blantyre District. This included the training ofMinistry of Health staff in the national guidelinesand protocols. This programme ceased in May 2003.

AgricultureGOAL has had two programmes in the agricul-

tural sector, both running during phases of JEFAP 1and 2. The first entailed distributing seeds and cut-tings of ‘non-maize crops’ to vulnerable rural fami-lies, to encourage crop diversification and the for-mation of community seed banks for sustainability.This programme, led by CARE and funded by theOffice of US Foreign Disaster Assistance (OFDA),involved training on seed production, multiplica-tion, storage and post harvest management, includ-ing food preparation techniques and organisationalcapacity building. This ran from late 2002 to June2004.

A second programme, funded by the FAO,involved working at two established governmentrun Nutritional Rehabilitation Units (NRUs), one inBlantyre and one in Chiradzulu. Based at the NRUs,GOAL trained two ‘Home Garden Managers’ inestablishing vegetable gardens and livestock pro-duction, as well as training ‘carers’ in home gardentechniques to support malnourished children. Inaddition, 1,500 families were supplied with an FAOstarter pack of seeds and tools to help support therecovery and sustain the health of the childrenattending the NRUs after their discharge. GOALinvolvement ended in May 2004 when the pro-gramme was handed over to a committee at each ofthe NRUs for future development and functioning.

Evolution of GOAL Activities in Malawi

Andy Nicholson is currentlyCountry Director of GOAL inMalawi. He has been in Africasince 1990 working mainly inemergencies in Sierra Leone,Liberia, Rwanda, DRC,Burundi and now Malawi. Hasworked for Lutheran WorldFederation, Christian Aid,Save the Children UK andnow GOAL.

The model farmers receive special training intopics such as coffee agronomy, post-harvesthandling, soil and water conservation, and inter-nal control systems. They are also provided withnecessary tools.

Project impactBy the end of the first year of the project

(January 2005), a number of outputs and impactshave been realised:

• Over 1,000 farmers (chronically poor and economically vulnerable) were trained in coffee agronomy, post-harvest handling, and soil and water conservation techniques.One example of a coffee improvment intervention was the provision of drying materials so that farmers could dry their coffee off the ground, thus improving coffee quality.

• Twenty farmer promoters (six women) weretrained in skills that could be transferred to their fellow farmers. Farmer groups were then formed around these promoters.

• Fifteen demonstration gardens, that are well located and easily accessible, have beenestablished.

• Over 30,000 coffee seedlings were distributed to farmers (chronically poor) to replacethe old coffee trees that had been affected by coffee wilt disease.

• A major focus is placed on linking farmers to appropriate market channels. The implemening partner, Ibero Ltd, bought all the coffee from the participants at a premium price.

• Eight communities developed proposals for agricultural training. The training included improved chicken rearing practices, horticultural training, and livestock managment. Additionally, some chronically poor households received chickens, goats, and piglets.

• Horticultural training was also conducted. Nearly 150 chronically poor households received direct inputs, such as vegetable seeds, fruit seedlings, and orange sweet potatoes.

• Over 600 participants were trained in improved knowledge and skills in nutrition,specifically for children and people living with HIV/AIDS. Out of these, 120 commu-nity nutritional trainers were identified and supported. Informational materials about child nutrition also were developed and distributed to the communities.

The futureKey issues identified throughout the project

will be addressed in the upcoming years.Seasonal changes and unreliable weather condi-tions have proved a particular challenge, sincethis delayed crop planting and so influenced thetiming of seasonal harvesting.

More attention is needed on the genderaspect of the project. Community awareness ses-sions could address the involvement of womenin coffee production and the production of othercash crops - particularly in the case of AIDS wid-ows who are supporting their children.Additionally, more interventions are neededspecifically to target and involve child-headedhouseholds in cash crop production.

Coffee production alone will not guaranteeimproved food security in households. For farm-ers with limited resources, food crop productionfor consumption and income generation remainsvital for attaining food security. Integrated farmmanagement, which does not focus on a singlecash crop, is needed to sustain the farm’s naturalresource base.

For further information, contact MosesKalyebara, email:moses.kalyebara@planinternational.organd Erin Culbertson, email: erin.culbertson@planinternational.org

By Andy Nicholson

The author would like to acknowledge his prede-cessor, Pat Mulcahy, and all the GOAL staff whohave contributed to GOAL’s work and achieve-ments.

FFW Community tree in Mbenje Nsanje

A N

ichols

on,

GO

AL

Field Article

This project is now being implemented byGOAL at two more NRUs in the districts ofNsanje and Chikwawa.

RehabilitationThe District Assembly has donated the for-

mer Presidential State House rent-free for threeand a half years for GOAL to use as an opera-tional base, in return for the investment made inrehabilitating the building.

APPLEFinally, GOAL commenced activities under

the EU funded Aids Prevention and PositiveLiving Programme (APPLE). This is focusing ontwo transport corridors, one wholly inMozambique, and the other, the Nacala corridor,running from Mozambique into Malawithrough Mangochi and on to Blantyre. This pro-gramme aims to provide support in developingan integrated HIV/AIDS Prevention, Care andSupport network along the corridor runningthrough the four districts of Machinga,Mangochi, Balaka and Blantyre. Two locationsfor Integrated HIV Prevention Care and Supportservices (IPCSs) have been established inBlantyre and Balaka. Both centres are now fullystaffed and training has taken place to build onthe capacity of personnel associated with theproject around the corridor, involving both min-istry of health staff and community volunteers.

A base line survey was completed at the endof February to establish the capacity of healthcentres and local community based organisa-tions, and to establish the levels of knowledge,attitudes and awareness on HIV/AIDS amongstprimary school and out of school youth withinthe operational area. Community orientationswith key leaders in each of the communitieshave also taken place. These meetings estab-lished how the communities are coping withHIV/AIDS, and the ways that HIV/AIDS can betackled, with an introduction as to how the EUApple programme can work with the communi-ty and health centres. GOAL has also equippedand opened three voluntary counselling andtesting (VCT) centres located within establishedhealth centres in the Blantyre district. The sur-prisingly large numbers of people received at allcentres on their first days of opening for testinghas been cause for great optimism.

For further information, contact AndyNicholson, email: andy.nicholson@goal-malawi.com, or GOAL Malawi,email:goal.admin@goalmalawi.com

FFW bridgeconstructionin Chiradzulu

A N

ichols

on,

GO

AL

Figure 1 Asset ownership categories by beneficiary status and gender of household head

24.5%

Asset ownership (different assets*)

Beneficiary household

Non-beneficiaryhousehold

% of total households

Total

Asset very poor (0-2 assets)

12.9%

M F M F M F

22.8% 27.6% 18.6% 25.9% 20.9%

Asset poor (3-5 assets) 45.2%

25.5%

46.4%

35.5%

6.6%

Asset medium (6-8 assets)

Asset rich (9+ assets)

100%

18.4%

Total 100% 100% 100%

37.4% 43.7% 37.7%

31.8% 22.3% 28.9%

38.7% 40.0%

29.1%

12.9% 10.6% 10.2%

100%

3.6% 10.3% 11.7%

100% 100%

44

This article was written based on a WFP consultationto Malawi in February 2005.

During 2003, WFP Malawi significantlystrengthened the monitoring activities for all its pro-grammes in Malawi. The Post DistributionMonitoring (PDM) and Community HouseholdSurveillance (CHS) programmes implemented byWFP are most relevant to targeting in Malawi andprovide invaluable insights regarding the role ofusing proxy indicators to target people living withHIV/AIDS (PLWHA).

Community Household Surveillance (CHS)The CHS is a regional initiative in six EMOP

(Emergency Operation) countries and was initiatedin early 2003. CHS is based on sentinel site monitor-ing and has four main objectives;

• to monitor impact of food aid• to monitor trends in food security• to monitor links between food security and

nutrition and/or HIV/AIDS, and• to feed into early warning system information.

The two instruments for the CHS are a house-hold questionnaire and a monthly focus group dis-cussion. CHS is conducted by WFP field monitors,and staff from the National Statistics Office (NSO),Ministry of Agriculture and Irrigation, and NGOimplementing partners.

CHS was initially piloted in Malawi in July 2003in 30 sentinel sites. The first round took place inOctober 2003, with 30 sentinel sites in six randomlyselected EMOP 10290 districts. Four FinalDistribution Points (FDPs) per district were select-ed, with one village randomly selected from servingFDPs. In each village, 22 households were selected,comprising 11 beneficiaries and 11 non-beneficiar-ies. A second and third round CHS were conductedin February 2004 and October 2004 respectively,with a fourth round planned for February/March2005.

The initial pilot CHS in July 2003 mainly collect-ed information on the food security situation,including data on coping strategies and household

perceptions on need for food aid. There was alsosome analysis of vulnerability. The first round ofsurveys collected and analysed data on food securi-ty and coping strategies but also focused on target-ing. The first round report claimed that social target-ing was successful, e.g. 37% and 27% female-head-ed households amongst beneficiaries and non-bene-ficiaries respectively. However, it stressed that the‘asset very poor’ were less represented than richerhouseholds (see figure 1).

The second report, in February 2004, concludedthat social targeting, i.e. female headed households,orphan containing households, elderly and dis-abled, etc, was successful, e.g. 41% of beneficiaryhouseholds were female headed compared to 37%non-beneficiary households, and 26% beneficiaryhouseholds had chronically ill compared to 21%non-beneficiary. It also stressed that 9% of benefici-aries had no social inclusion criteria, while 31% ofnon-beneficiaries had three or more social inclusioncriteria. Furthermore, 54% of beneficiaries wereasset poor, compared to 48% for non-beneficiaries.

Post Distribution Monitoring (PDM)The main objectives of PDM are to monitor the

use of food aid, satisfaction with food aid, andaccess to food aid (involving targeting inclusion andexclusion). It involves a household questionnaireand focus group discussions and is conducted byNGOs and WFP field monitors. The first PDMreport was compiled in December 2003, containinginformation for September – December 2003 (EMOP10290 began in July 2003). There had been consider-able work in the development of the PDM prior tothis, as well as evaluation of its implementation. Aconsultant on monitoring and evaluation was takenon at country office level, and experiences of PDMsfrom other countries with similar programmes weredrawn upon.

The first PDM narrative report was based on atotal of 1196 households, with a relatively equal dis-tribution between beneficiary and non-beneficiary.Beneficiary households were those receiving foodaid through the new EMOP under Food for Work(FFW), or Vulnerable Group Feeding (VGF), i.e.those infected or affected by HIV/AIDS, children,orphans, expectant and nursing women. Social vul-

WFP Monitoring and Evaluation of HIV/AIDS Programming in MalawiBy Jeremy Shoham, ENN

*excludes livestock

Evaluation

Post distribution monitoring GOALfood distribution Chiradzulu 2

A N

ichols

on,

Goal

45

social targeting may have been over-empha-sised, i.e. the differences found between benefi-ciary and non-beneficiary groups were not sta-tistically tested and almost certainly not of suffi-cient margin to make certain claims.Furthermore, there seems to have been anundue emphasis on inclusion criteria ratherthan exclusion criteria, which showed very highlevels reflecting the large proportion of verypoor and poor in Malawian communities. It canbe argued that exclusion errors are more impor-tant than inclusion errors, because they are divi-sive at community level.

The monitoring has a number of methodologi-cal weaknesses:

• Comparing assets between beneficiaries and the non-beneficiary population is con-ceptually flawed, as one would expect non-beneficiaries to begin disinvesting in an emergency so that asset ownership would begin to equalise during the programme.

• Inclusion and exclusion criteria are normal-ly calculated on the basis of proportion of beneficiaries out of total target population, rather than proportion of beneficiaries who meet inclusion and exclusion criteria. The adopted method may lead to an underesti-mate of targeting error.

• The differences between beneficiary and non-beneficiary inclusion/exclusion percentages are not compared statistically.

RecommendationsGiven the lack of success of economic target-

ing, there is clearly a need for more piloting andresearch to determine when and where such tar-geting is feasible and appropriate. For example,where a large proportion of the population areeconomically ‘poor’ or ‘very poor’, it may not befeasible. It may be that in such circumstances,social targeting is more acceptable at communi-ty level while economic based criteria will leadto conflict and disagreement. An implication ofthis may be that ‘the first cut’ of targetingshould be social targeting, with economic tar-geting subsequently invoked once these house-holds have been identified at community level.In order to assess efficiency of economic target-ing, it would have been better to carry this out atthe onset of the programme or following newregistrations as, during the programme, aprocess of equalisation is likely to occur andnon-beneficiaries are forced to disinvest. Wherepossible, statistical tests should be applied. Itshould have been possible to apply non-para-metric statistical tests (e.g. pearson coefficients)to compare social inclusion /exclusion for bene-ficiary and non-beneficiary populations. Hadthis been done, the findings would have beenmore credible.

For further information, contactJeremy Shoham, email:jshoham@easynet.co.uk

1 The percentage of beneficiary households with access to live stock decreased from 30% to 21%. However, there was no improvement with regard to access to land.

2 The analysis considers eight major exclusion indicators proposed by the JEFAP monitoring working group. These relate to livestock, acres of food crop and cash crop land, income as medium business and income as formal wage. There are six inclusion socio-vulnerability indica-tors. These are single headed household, more than seven persons in a HH, elderly headed, female headed, keeping orphans and keeping chronically ill.

3 In order to assess the targeting efficiency, a targeting index is derived for each household, which is composed of the difference between its number of inclusion and exclusion criteria. For the purpose of simplicity, each exclusion indicator is given a negative one (-1) and eachinclusion indicator a positive one (+1), When the two are aggregated, a deserving household is one that has apositive difference, i.e. more inclusion indicators than exclusion indicators. A negative targeting index will mean that the household has more exclusion criteria than inclusion criteria and hence could be considered as food secure.

4 It should be noted that there are different figures in the text of the report for inclusion and exclusion and effi-ciency (16% and 73% and 78% respectively). Joint Emergency Food Assistance Programme, a country wide consortium of 12 international and local NGOs in Malawi

nerability categories under this EMOP includedfemale headed households (FHH), elderly headed,households with disabled member or orphans orthe chronically ill (CI), or households with a highdependency ratio. Social indicators were combinedwith food security indicators, i.e. access to land,access to livestock, access to regular income or assetownership.

The main findings of the first report were thattargeting on the basis of social vulnerability wasgood but weak with respect to economic wealthindicators, i.e. overall access to livestock, income orland was very similar between beneficiaries andnon-beneficiaries. This was believed to be due to ahigh level of inclusion errors. The report recom-mended refining inclusion criteria through system-atic discussions with local communities and tradi-tional chiefs, giving the opportunity to work outlocal checklists of social and economic indicatorstailored to a given local context.

The January 2004 PDM report found a clearimprovement of targeting with respect to social andeconomic welfare indicators following a beneficiaryverification exercise in late 20031 . The February 2004PDM report stated that the VGF programme morefrequently included households keeping orphans(73%), the chronically ill (31%) and FHH (24.4%)compared to FFW activities (43%, 19% and 43%respectively). It concluded that VGF targetingremains satisfactory with a predominant inclusionof beneficiary households with at least one socialvulnerability criteria.

The March 2004 PDM introduced new analyticalvariables – overall inclusion and exclusion errorsand targeting efficiency2. The executive summary ofthe report states that inclusion error is 16% whileexclusion error is 30%, and calculates that the target-ing efficiency is 78%3. The June PDM found an inclu-sion error of 18% and exclusion error of 54%. Thusthe targeting efficiency was 64%4.

Informing initiativesThe PDM has had a significant impact on the

targeting system in EMOP 10290 and the currentProtracted Relief and Recovery Operation (PRRO).Early findings, after the first report, led to an evalu-ation of targeting (Nsama report in March/April2004) and an evaluation of the HIV/AIDS pro-gramme, with a particular emphasis on targeting inApril/May 2004 (Selaphera Consulting Ltd).

Findings from the PDM and CHS have also ledto a number of revised monitoring initiatives,including where:

• Implementing partners were strongly encour-aged to undertake systematic and re-iterated targeting verification exercises.

• In May 2004, a collaborative process was initiated between WFP and the JEFAP5 co-operating partners to review the targeting criteria. As a result, a final set of inclusion and exclusion criteria was thoroughly discussed, agreed, and later on included in the revised implementation guidelines (JEFAP III).

• Implementation of the new set of criteria and plans to appraise the new set of inclusion and exclusion criteria.

ConclusionsAlthough there was no rigorous monitoring of

targeting during the initial EMOP (1200), partly dueto the need to prioritise implementation of the gen-eral food distribution, the monitoring evolved rap-idly into an extensive system which collected a vari-ety of data needed to assess compliance with target-ing. The system has compiled a large quantity ofextremely useful data. The level of sophisticationachieved with regard to monitoring targeting isunique and could serve as a model for other pro-grammes.

The findings of the monitoring, in particular theweaknesses of targeting on the basis of economiccriteria, were rapidly identified and reflected inJEFAP 111 guidelines.

Some of the claims regarding the success of

The HIV/AIDS epidemic in Malawi is receiv-ing increasingly committed responses fromits government, donors, religious institu-tions, the UN, and NGOs. Programmes

addressing prevention, testing, treatment and vulner-ability are gaining momentum. Currently many ofthese programmes aim specifically to targetHIV/AIDS infected/affected individuals and/orhouseholds. While this may be an appropriateapproach for many types of interventions (such asHIV-specific prevention, curative, medical, educa-tional or nutritional programmes), for some foodsecurity inputs programmes this may not be the mosteffective approach. The majority of rural Malawiansare chronically impoverished and malnourished;they exist on the edge, where a shock to, or within, anaverage rural household can push it to a degree ofvulnerability that they may not be able to withstand.

Within this context, it may be more important toimprove the selection of the most vulnerable house-holds, whether the vulnerability is rooted inHIV/AIDS or not - first considering the degree andaspect of a household’s vulnerability, and then con-sidering the HIV status of the beneficiaries whendeveloping the most appropriate intervention. Byaccurately targeting the more vulnerable householdsin countries, like Malawi, with high rates of HIVinfection, infected/affected households will neces-sarily be targeted.

Action Against Hunger (AAH) examined theweaknesses and strengths of the current proxies usedto identify HIV/AIDS infected/affected and/or vul-nerable households in rural Malawi. After identify-ing the drawbacks of the proxies, AAH developed amethodology to mitigate their weaknesses. Themethodology was piloted during an ECHO

TargetingVulnerableHouseholds Within theContext ofHIV/AIDS in MalawiSummary of evaluation1

By Maja Munk and Dr. Neil Fisher

Maja Munk has been working in the food securitysector for ACF/AAH in Africa and Asia since 2002.Her photographs have been published in periodicalsand books in the Americas, Europe and Asia. She iscurrently working in Liberia as a programme coor-dinator.

Since 2002, Neil Fisher has been working in Malawifor AAH and is currently responsible for theIntegrated Nutrition and Food Security SurveillanceProgramme. Previous work includes 12 years withuniversities in Kenya and Nigeria teaching CropProduction and Farming Systems and researchingindigenous farming systems.

Evaluation

46

(European Commission Humanitarian Office) fund-ed farm inputs distribution in Malawi’s central dis-trict of Ntchisi.

In Malawi, like many other countries, there isfear, ignorance, a lack of accessible free and anony-mous testing, and a common assumption that all peo-ple tested for HIV have AIDS. Many women needtheir husbands’ approval to be tested, and there isdifficulty preserving privacy within the village con-text. Given the personal and social ramifications ofknowing your status and/or being associated withthe disease, few rural Malawians undergo voluntarycounselling and testing (VCT) or are willing to beopenly identified as being HIV positive. Until theimpact of the current investments in VCT and anti-retro viral (ARV) treatment are felt by Malawi’s ruralpopulation, and the rural population is successfullysensitised to HIV/AIDS, identifying HIVinfected/affected household will continue to be prob-lematic. Additionally, there may be negative ramifi-cations for households associated with programmesspecifically targeting HIV/AIDS infected/affectedfamilies.

Correctly targeting vulnerable households is dif-ficult in most contexts; accurately targeting house-holds who are vulnerable directly or indirectlybecause of HIV/AIDS is, on any significant scale, notpractically possible in rural Malawi. The substantialincrease of VCT and ARV throughout the countrymay, in the future, make correct targeting of thesehouseholds possible. However, currently donors andNGOs are increasingly attempting to address theHIV/AIDS pandemic by targeting HIV/AIDS infect-ed/affected households. In the absence of knowledgeof an individual’s HIV status, proxies are often used.The popular proxies, while theoretically indicative,are often in practice imprecise, and may not, inMalawi’s context, be the most effective tools to selectthe HIV/AIDS infected/affected vulnerable popula-tion.

ProxiesIn Malawi, there are commonly used proxies

(such as mortality /morbidity, orphan, female/child /elderly-headed households, and high dependency ratio households) that are associated withHIV/AIDS and vulnerability. For a variety of reasons,these proxies may be problematic.

Common proxies‘Chronic illness’ is a popular proxy often associ-

ated with HIV, but the term is often not well defined.Sometimes ‘chronic’ is given a time scale (i.e. threemonths, one year, etc.), but commonly the definitionof ‘illness’ is left assumed, rather than defined.Conversations with villagers indicate that peoplehave a very varied idea of what ‘chronic illness’means - even within a given timeframe. At the villagelevel, a chronically ill person can be someone suffer-ing from a bad back or arthritis, or someone who hashad several bouts of different related or non-relatedailments within the specified time. While a well-defined ‘chronic illness’ may capture householdmembers who have been unhealthy for a specificperiod of time, it can be an imprecise proxy for cap-turing HIV/AIDS or its opportunistic illnesses. Thereare also many types of chronic illnesses historicallyexisting in southern Africa that may not have anyrelation to HIV/AIDS or its associated diseases.However, ‘chronic illness’ may be a good indicator ofhousehold stress due to a possible decreased labourcapacity and/or increased expenditure.

The recent death of a household member is also acommon proxy. Unlike ‘chronic illness’, ‘death’ is

well defined. Historic data support the view that thepresence of HIV/AIDS has lowered Malawi’s lifeexpectancy. However, one must take into account theperson’s age and reason for death before the death isassociated with HIV/AIDS. While the presence ofdeath does not necessarily indicate the presence ofHIV/AIDS, if the person who died was a productivemember of the household, ‘death’ may indicateincreased household stress.

Like chronic illness, the term ‘orphan’ lacks a pre-cise definition. While NGOs commonly assume anorphan to be a child with one or both parents dead, atthe village level the definition of ‘orphan’ is not soclear-cut. Numerous individual and group discus-sions revealed that central rural Malawians have awide spectrum of opinions when it comes to definingan orphan. The varying definitions fall into two basicschools of thought: an orphan is a child who has oneor two parents dead irrespective of the support thechild receives from its remaining family; or anorphan is a child who does not have access to basicnecessities, whether its parents are alive or not, andmay rely on people other than its parents to satisfythose needs. Even within a household, there can bedifferent options regarding the status of a child. Awoman’s second husband may consider the childrenfrom his wife’s first marriage to be orphans, the wifemay or may not consider her own children asorphans.

If an orphan is defined by the death of one orboth parents, there is also the question: once anorphan always an orphan? That is, if a child is consid-ered an orphan because one parent died, does itremain an orphan after the remaining parent remar-ries? Does an orphan continue being an orphan afterthey get married or on their 16th or 18th birthday? Oris being an orphan restricted to a state of beingdependent on others for your needs? If a grandmoth-er is taking care of her grandchild, how much sup-port is the child’s parent/s responsible for, before thechild is considered an orphan? While the answers tothese questions may, on an abstract level, seemstraightforward, at the field level they are often com-plicated.

The presence of ‘orphans’ in a household is notnecessarily a reliable proxy for HIV/AIDS in Malawi.While governmental and non-governmental actorshave developed definitions for ‘orphans’, often thesedefinitions do not translate into field realities. Itshould not be automatically assumed that the pres-ence of an ‘orphan’ necessitates a death in the child’sfamily (as one, both or neither of the child’s parentsmay be dead), or that the presence of the ‘orphan’necessarily indicates a long-term economic burdenon the hosting family (as the child may be receivingother support, the presence of an older ‘orphan’ mayrepresent a positive or neutral contribution to ahousehold, or the child may be hosted for a shortperiod of time). However, the hosted orphan may beat a higher risk of vulnerability due to inequality ofcare, than the other children in the household. Thatsaid, specifically targeting the orphan, as oppose tothe household as a whole, may further marginalisethe child.

The proxy ‘female-headed household’, asopposed to male-headed, assumes the absence of anactive adult male contributing to the house. Theabsence of a male-head has a compounded effect: itputs more of a burden on the female-head as the pri-mary provider for the household, often in an environ-ment where women have fewer income generatingopportunities than men. AAH surveys in centralMalawi have found that polygamy in rural centralMalawi is quite widespread. However, the scope ofpolygamy does not readily present itself, as thepolygamy practised in central Malawi is matriloca-tional, i.e. the male-head divides his time betweenthe households of his co-wives, as opposed to sharinga common compound with them. In this structure,each co-wife’s house lacks the full contribution of thehusband or the support of the other co-wives.Interviews with co-wives found that these house-holds can effectively be female-headed, as the wifehas to provide for the household with, most com-monly, little or no support from the husband.However, these households can present themselvesas male-headed as there is a male head of house, evenif only figuratively.

Additionally, in a culture where the man controlshousehold expenditure, a male’s presence can benegative if the priorities of the man contradict the pri-orities of the household’s food security - a commoncomplaint of village women. Therefore, it is possiblefor a household to be male-headed but the male’scontribution to the household to be neutral or nega-tive. What may matter more than the sex of the headof house is the number of active contributing adults,and to what degree they positively contribute to thehousehold.

‘Elderly’ and ‘child-headed’ households are like-ly to be vulnerable as they are often without a fullyactive contributing adult. The degree of their vulner-ability lies both in the household head’s ability tocontribute to the house and the amount of supportthat they receive from their social-safety network.This can be a particularly important resource for eld-erly-headed households. Often the elderly live inclose proximity to their adult children, however, thesupport provided can vary widely. Child-headedhouseholds may have to rely on a smaller family sup-port structure (as they have no adult children to relyon). It is also important to remember that the thresh-old for childhood varies culturally in Malawi, it is notuncommon for teenagers to marry and establish theirown households before their 18th birthday. Whilethese households may be technically child-headed, tocall them such may be misleading. While elderly-headed households with children are frequentlyassociated with HIV/AIDS, the presence of a genera-tion gap in a household is not unusual where middleage groups migrate for work. This predates the onsetof HIV/AIDS and should not automatically be asso-ciated with HIV/AIDS. However, as the family sup-port structure in Malawi shifts focus from the extend-ed family network towards the nuclear family, childand elderly-headed households may become increas-ingly disenfranchised.

Developing alternative proxiesAlthough measures are underway to step-up

VCT, widespread HIV testing of rural Malawians willremain problematic in terms of ethics and plausibili-ty. Therefore, developing empirical proxies is neces-sary. The reason why HIV/AIDS infected/affectedhouseholds are targeted is because the presence ofthe disease is believed to increase vulnerability.Therefore, one way of indirectly identifyingHIV/AIDS infected/affected households is to identi-fy a household’s degree and aspect of vulnerability.

Calculating a household’s dependency ratio is acommon way to determine a household’s degree ofvulnerability. In the classic dependency ratio, thenumber of dependent members is expressed as aratio to the productive members (DependencyRatio=100*number of dependents/number of pro-ductive adults). The problem with this equation isthat a person is valued as all (productive) or nothing(dependent). This rating of contribution does notreflect the gradient of contribution over a lifetime.

AAH explored an alternative measure of depend-ency that allows for varying degrees of contribution,and developed a ‘food/labour dependency ratio’.Knowing the age and gender of each householdmember, one can calculate the theoretical daily foodenergy need and potential labour capacity of a house-hold.

When calculating the potential contribution thatan individual makes towards generating cash and/orkind, an adult (21-55 years) male was rated at 1.0 (ashis workload is primarily limited to income earningand/or crop production) and an adult female at 0.8(as her workload consists of not only income earningand/or crop production, but also household work,child rearing, caring for the ill, etc.). The positive con-tribution of children starts at 12 years old (rated at 0.1for males and 0.08 for females) and increases until 20years, when they become a full adult. Then their con-tribution decreases from 56 years of age until 68 yearsof age (when they are valued at 0.09 for males and0.02 for females), after 68 years they are rated at zero.

Similarly, the food energy requirements of eachhousehold member is calculated on a sliding scale bysex and age, starting at 1300kcals for female and maleone-year olds and ascending to 2900kcal for lactatingactive adult females and 3300kcals for active adultmales. By dividing a household’s food energy

Evaluation

A woman receivesgroundnuts as partof the programme

N F

isher

, M

alaw

i

requirement by its labour potential, the ‘food/labourratio’ is formed. At 5,800 kilocalories per productivemember, an average family of two active and threeinactive members in Malawi is generally food secure.As the active household members become responsi-ble for more than 5,800 kilocalories, the householdbecomes increasingly stressed.

Targeting in Ntchisi districtBeneficiaries were selected on the basis of house-

hold food/labour ratio and degree of material pover-ty. Chronically ill adults2 were assumed not to be con-tributing to the labour of a household. Dependencywas then scored from 1 (lowest) to 10 (highest). Thepoverty score was created by placing a monetaryvalue on productive and non-productive assets andcalculating their total value. The total was then divid-ed by the number of people within the household toformulate a per-capita value, and ranked from 1(poorest) to 10 (richest). The dependency/povertyscore was then created:

10+Dependency Score–Assets Score=Dependency/Poverty Score

The score took values from 1 (least vulnerable) to19 (most vulnerable). After this score was ascer-tained, one point was added for each orphan3 in thehouse. Two points were added to households thathad elderly people caring for children. Householdsthat scored 11 or more points were chosen as benefi-ciaries.

A short, ten-minute, targeting questionnaire cap-turing the essential data needed (household struc-ture, mortality/morbidity, assets, and land holdings)for the targeting methodology was conducted in fourhealth centres and their surrounding villages, withinthe catchment. Of the 3,500 people interviewed withthe questionnaire, 64% of the total number of house-holds qualified under the criteria used in the target-ing methodology. Due to time constraints, theremaining 36% of beneficiaries were selected throughlocal leaders using defined criteria: female-headedhouseholds, elderly-headed households with chil-dren, and households with orphans. Household char-acteristics determined the inputs received: vulnera-ble households with sufficient land and labourreceived crop inputs, vulnerable households withinsufficient land or labour received chickens andfeed.

Evaluation of the methodologyConstraints

The effect of the piloted targeting methodologywas diluted, and AAH’s ability to evaluate it wassomewhat limited, as, due to time constraints, not allbeneficiary households were targeted through thefood/labour dependency ratio. Additionally, AAHliased with local leaders to inform selected house-holds and ensure that they collected their beneficiarycards at designated points before the distribution.Follow-ups found that some of the cards were givento people other than the intended beneficiaries.

During the monitoring and evaluation process, itwas not possible to determine under which method-ology a particular household was targeted. It was,therefore, not possible to ascertain during the evalu-ation process if incorrectly targeted households wereselected by the piloted methodology or selected bylocal leaders. As an unknown third of the beneficiar-ies were not selected through the piloted methodolo-gy, properly evaluating its ability to correctly identi-

fy vulnerable households was difficult.

Selected householdsOf the surveyed beneficiary households, 65%

qualified on one criteria count and 17% qualified ontwo criteria counts. 18% should not have qualified asthey were in the richest two quartiles and in the twoquartiles with least dependency.

Despite the fact that the potential effectiveness ofthe food/labour dependency ratio may have beendiluted as 36% of the beneficiaries were selected vialocal leaders, the methodology was able to skew thetargeted households towards the higher food/labourdependency households when compared to the basepopulation. The only exception was the proportion ofbeneficiaries selected in the >8000kcal group, whichis slightly less than the base population. This under-representation may be caused by these disenfran-chised households not accessing health structuresand not having a voice in the community, and there-fore less likely to be picked-up during the selectionprocess.

In the food/labour ratio (figure 1), the thresholdfor vulnerability is calculated at 5800kcal per manequivalent. The targeting methodology was able toselect proportionally more households that have ahigh food/labour dependency. In the dependencyratio, household vulnerability increases as the ratioincreases from 100 -the methodology was able to pro-portionally select more of these high dependencyhouseholds as well (see figure 2).

Households which were not selected by the ques-tionnaire had a food/labour ratio of 4,825kcal perman equivalent. The 17% of targeted householdswho should not have qualified had a food/labourratio of 5,056kcal per man equivalent. Qualifyinghouseholds had a food/labour ratio of between 6,321and 8,000Kcal per man equivalent.

There were statistically significant demographicdifferences between targeted households and thebase population. Compared to the base population,the piloted targeting methodology picked up propor-tionately more households headed by persons aged31 years or older, headed by women, and/or headedby widow/ers. These households had fewer produc-tive and non-productive assets than the base popula-tion. The presence of orphans in targeted householdswas between 2% and 4% higher than the base popu-lation. The poorest and most dependent 17% of thesample was more likely to be female-headed house-holds with orphans; if these households had orphansthey were more likely to have a higher number oforphans per household than the other segments ofthe sample.

Estimates of inclusion/exclusion errorsThe piloted methodology had an estimated inclusionerror of between 21% and 38% and exclusion error of<49%. The inclusion errors appear to be better thanthose of the government of Malawi’s DFID4 fundedTargeted Input Programme (TIP) in 2000-01. Thereare a number of possible reasons for this large inclu-sion error: the 36% who were selected through localleaders, rather than through the targeting interview,may not have conformed to the dependency/pover-ty criteria that were used for the 64% selected by thepiloted methodology; the correct people were inter-viewed for the beneficiary profile survey, but theygave different answers to the household structureand asset questions during the survey than duringthe targeting interview5 ; some beneficiary cards were

47

misallocated. If these assumptions are true, then thetrue inclusion error lies somewhere between 21% and38%, as the portion attributed to households givingdiffering answers is, in part, a measurement errorrather than a real inclusion error. While the inclusionerror of the piloted methodology is the same or worsethan the inclusion error for AAHs distribution inNtchisi’s neighbouring district of Kasungu (see table1), where a traditional method of targeting throughlocal leaders was used, the exclusion error is better.This is important as exclusion errors are arguablymore important than inclusion errors, as they cantrigger resentment within the concerned communi-ties.

The quantitative inclusion and exclusion errors,as well as the qualitative input of the enumeratorsand AAH food security staff who carried out theevaluation, indicate that within Malawi’s context ofwidespread endemic poverty, it is difficult to differ-entiate the subtle degrees of poverty or dependencyamong poor households to find the most vulnerableamongst the vulnerable. However, as most interven-tions do not have the capacity or mandate to target allof Malawi’s impoverished households, selection cri-teria are needed. While the methodology piloted inNtchisi is a step in the right direction, more subtletargeting tools will need to be developed. While iden-tifying the 20 to 30% “least poor” is reasonably sim-ple, differentiating between levels of poverty in theremaining 70 to 80% is much more difficult.

ConclusionsIt is not HIV/AIDS itself that threatens a house-

hold’s livelihood, but the ramifications of the diseasethat make the household increasingly vulnerable astheir income and expenditure are skewed, and as theratio of active members to dependents shifts. Perhapswithin Malawi’s context of widespread chronicpoverty, it is less essential for NGOs implementingnon-HIV specific food security interventions to findhouseholds infected/affected by HIV/AIDS than toidentify households that are vulnerable. The natureof the vulnerability should be taken into accountwhen designing programmes. Once the most vulner-able households have been correctly identified, thespecific stresses of HIV/AIDS can be considered. Dueto the limitations of proxies being applied practicallyin rural Malawi, humanitarian actors should not usethem as the simple solution to identify beneficiaryhouseholds and should question if using them is, infact, appropriate for their intervention. While the tar-geting methodology developed by AAH was a for-ward move, methodologies to improve identificationof the most appropriate vulnerable households to tar-get should be further developed and explored.

For further information, contact AAH Malawi,email: aah@globemw.net

Table 1 Estimates of inclusion/exclusion errors in AAH programmesand other programmes

Figure 1 Food/Labour Ratio of the Population and Beneficiaries

Figure 2 Classical DependencyRatio of the Population and Beneficiaries

Intervention

AAH Ntchisi crop-beneficiaries (2003-04)

AAH Ntchisi chicken-beneficiaries (2003-04)

Exclusion Error %Inclusion Error %

21 to 38

Less than 27

Targeted Input Programme (2000-01)

AAH Kasungu beneficiaries (2003-04) 21 Less than 55

35 59

General Food Distribution (2002-03)(AAH/GoM surveillance data)

62 41

General Food Distribution (2002-03)(Nyirongo et al6)

70 26

Less than 49

1 Munk M and Fisher N (2004). Targeting vulnerable householdswithin the context of HIV/AIDS. An evaluation of a piloted methodology. Ntchisi, Malawi. Action Against Hunger

2 For the questionnaire, ‘chronically ill’ was defined as ill for more than three months

3 For the questionnaire, orphan was defined as under 18 years old with both parents dead

4 UK Department for International Development5 Targeting interviews were not conducted in the home, impact

assessment interviews were6 Nyirongo CC, Msiska FBM, Mdyetseni HAJ, Kamanga FMCE,

Levy, Sarah (2003). 2002-03 Extended Targeted Input Programme: Evaluation module 1: Food production and security in rural Malawi. Malawi Ministry of Agriculture, Irrigation and Food Security and UK Department for International Development.

Evaluation

The Emergency Nutrition Network (ENN)

grew out of a series of interagency meetings focus-ing on food and nutritional aspects of emergencies.The meetings were hosted by UNHCR and attendedby a number of UN agencies, NGOs, donors andacademics. The Network is the result of a sharedcommitment to improve knowledge, stimulatelearning and provide vital support and encourage-ment to food and nutrition workers involved inemergencies. The ENN officially began operations inNovember 1996 and has widespread support fromUN agencies, NGOs, and donor governments. Thenetwork aims to improve emergency food and nutri-tion programme effectiveness by:

• providing a forum for the exchange of field levelexperiences

• strengthening humanitarian agency institutionalmemory

• keeping field staff up to date with currentresearch and evaluation findings

• helping to identify subjects in the emergencyfood and nutrition sector which need moreresearch.

The main output of the ENN is a tri-annual publica-tion, Field Exchange, which is devoted primarily topublishing field level articles and current researchand evaluation findings relevant to the emergencyfood and nutrition sector. The main target audience of the publication are foodand nutrition workers involved in emergencies andthose researching this area. The reporting and ex-change of field level experiences is central to ENNactivities.

The Team

Editorial team

Deirdre HandyMarie McGrathJeremy ShohamRupert Gill

Design

Orna O’Reilly/Big Cheese Design.com

WebsiteJon Berkeley

Contributors for this issueFiona MitchellAndy NicholsonPeter Paul IguMargarethe JunckerErin CulbertsonMoses KalyebaraKarine GarnierRuth SitumuBen WatkinsRobert AdangakuluStacia NordinPamela FergussonShannon SenefeldVictor Ochieng OwinoAmador GómezAntonia TorreblancaEileen KimTed Greiner

Pictures acknowledgement

Many thanks to all those, too numerous tomention individually, who kindly posed forpictures for Mary Corbett on her trip, or whosupplied ENN with pictures to use in thisspecial issue.

On the coverThe lady in the picture is the late MamaChristine with her youngest son (both HIVpositive). Apart from her own 5 children,Mama Christine was also caring for threechildren of her late brother-in-law. Since Mama Christine passed away about oneyear ago, her children have been living bythemselves, under the supervision of REEP.The oldest son is in secondary school(sponsored by REEP).

The Emergency Nutrition Network is a companylimited by guarantee and not having a share capital.Registered in England and Wales number: 4889844Registered address: Unit 13, Standingford House,Cave Street, Oxford, OX4 1BA, UKENN Directors: Jeremy Shoham, Marie McGrath.

Field Exchangesupported by:

Dan George is the ENN financeassistant, working part-time inOxford.

Rupert Gill is ENN administratorand project coordinator, based inOxford.

Jeremy Shoham (Field Exchange technical editor)and Marie McGrath (Field Exchange production/assistant editor) are both ENN directors.

This publication was made possible throughthe support provided to the Food and NutritionTechnical Assistance Project (FANTA) by theOffice of Health, Infectious Disease andNutrition of the Bureau for Global Health atthe U.S Agency for International Development(USAID), under terms of CooperativeAgreement No. HRN-A-00-98-00046-00awarded to the Academy for EducationalDevelopment (AED). The opinions expressedherein are those of the author(s) and do notnecessarily reflect the views of USAID.

Susan ThurstansMary CorbettKathyrn LockwoodMartin Davidson MtikaRichard MmangaSaul GuerreroPaluku BahwereKate SadlerSteve CollinsJeremy ShohamMaja MunkNeil FisherGertrude KaraMrs Nellie Nyang’waAndrew MellenKate VorleyMieke MoensMary Makopkha

48

Emergency Nutrition NetworkUnit 13, Standingford HouseCave Street, Oxford, OX4 1BA, UK

Tel/fax: +44 (0)1865 722886Email: office@ennonline.netwww.ennonline.net