Learning from nutrition interventions in Eritrea, Ethiopia and Kenya

Summary of evaluations¹

SC UK survey team in Eriteria

SC UK recently evaluated a number of emergency nutrition responses undertaken in Eritrea (Gash Barka and Northern Red Sea), Ethiopia (Legambo and Fik) and Kenya (Wajir) during 2000-2001. Reviewed programme interventions included targeted and blanket supplementary feeding, and therapeutic feeding.

Whilst the content of the main recommendations from the evaluations are not ground breaking, they do reiterate the fundamentals of good programming, and the importance of respecting these principles in the difficult circumstances in which emergencies often present. In two of the contexts considered (Wajir and Fik), the affected pastoral populations presented specific programme challenges, while three of the contexts were affected by insecurity (Wajir, Fik and Eritrea). Moreover, since the emergencies coincided in terms of timing and region affected, considerable pressure was placed on SC UK's capacity to recruit additional staff and provide necessary programme support.

The following brief highlights of the key lessons learned may prove useful to other agencies operating in similar contexts.

Preparation and decision-making

From the outset, investment must be made in establishing management systems, such as finance, logistics and administration. Failure to do this reduces the effectiveness of technical staff whose energies are diverted to administrative issues, and risks compromising both programme effectiveness and accountability to donors.

Integration of therapeutic feeding centres (TFCs) into existing health structures such as hospitals, increases efficiency, is more economical and builds upon the skill levels of Ministry of Health (MOH) staff.

Where a programme is predominantly nutrition focused, health outcomes can sometimes be poorly defined. The health component of the nutrition work must be clearly specified with measurable objectives. Particular attention should be given to ensure that reasonable vaccination coverage is achieved in order to prevent further nutritional deterioration. Mass health education, as well as individual advice and treatment, can also be built into a nutrition programme.

In pastoral areas, fixed feeding centres in central locations may not be effective. In such settings, it may prove difficult to conduct cluster sample surveys or to alert people in advance when distributions are to occur. Also, people are not necessarily able to remain in one place for a long period of time. For these reasons, alternatives need to be explored, such as community based therapeutic feeding. More resources and attention are needed to monitor and supervise feeding programmes in such an environment.

Difficulties in recruiting experienced international staff need to be addressed. This could be achieved through developing an apprenticeship scheme for on-job training under the guidance of a mentor. Also, guidance on rapid, but fair and transparent, recruitment processes for national staff in an emergency is particularly important when opening new operational bases. This helps to reduce future staff management problems and ensure recruitment of the best available staff from the outset. Staff with local knowledge can be critical in managing activities such as staff recruitment, and in dealing with local matters, e.g. clan issues and security.

Nature of interventions

Criteria for withdrawal from a country or area must be identified from the outset of operations, and the programme reviewed against these on a regular basis. These criteria, and the strategy for withdrawal, may need to be revised but will provide a benchmark against which the programme can be reviewed.

Realistic timeframes for interventions should be established from the beginning. This makes for greater continuity of staff and allows for more effective planning. In particular, clear and consistent messages can be given to partners and local authorities about programme objectives, criteria for withdrawal and expected length of stay.

High turnover of international staff, particularly project managers, places programmes under considerable strain and can lead to:

• lack of continuity and consistency in management
• inevitable gaps between international staff assignments, which lead to unsatisfactory interim arrangements for project management
• poor relations and rapport with project partners
• recruitment demands on programme management.

Wherever possible, the preparation and implementation of national guidelines for nutrition in emergencies should be promoted, thus investing in intervention sustainability. Guidelines need to be updated regularly.

Issues to consider regarding supplementary feeding programmes (SFP's) include:

• Staff numbers must be based on the number of beneficiaries and the geographical coverage of the programme. The location of a SFP needs to take into account accessibility for the population and beneficiary numbers, which in turn will affect coverage and defaulting rates.
• Purpose built structures are resource intensive and limit the opportunities for capacity building among MOH staff. Existing health facilities can be used to assist health aspects of the programme, but travel implications for beneficiaries and additional demands on health centre staff may reduce the quality of the programme.
• Careful consideration should be given before opening a SFP in areas without an adequate general food ration (GFR). However, the reality is that SFPs will be set up even where there are no resources for a GFR, or the GFR is inadequate. In these instances, the supplementary ration should take into account the likelihood of the ration being shared.
• Blended foods must be fortified to ensure adequate supplies of micronutrients.
• Targeting pregnant women during the last six months of pregnancy may prove problematical, as determining gestational age is time consuming and may not be practical.
• In general, a comprehensive SFP should be up and running prior to a TFC being established. Supplementary feeding reaches many more children and vulnerable groups and should help prevent children from becoming severely malnourished.

Nutrition surveys can be used to establish the effectiveness of interventions and provide the stimulus for programme improvement. However, surveys can be costly and time consuming exercises, and may be particularly difficult to carry out in pastoral communities. Nutrition surveys should only be implemented where results will directly inform decisions about the type and scale of response. Furthermore, surveys should be standardised to allow comparison over time and between regions, and should gather relevant information on food security and health. Where possible, quicker and cheaper ways of making decisions should be explored. For example, SC UK is currently exploring the possibility of using lot quality assurance sampling².

Ensuring community involvement in establishing feeding programmes helps in planning issues, such as:

• Appropriate siting of centres.
• Understanding and support of feeding programme objectives, including acceptance of targeting criteria. For example, the rationale for admission according to anthropometric criteria, irrespective of socio-economic status, may be difficult for communities to understand without proper explanation.
• Participation of mothers/ carers throughout the length of stay in a feeding programme.
• Reduction in defaulter rates by carers understanding how the feeding programme works, better siting of centres, and advice on population movements.
• Providing volunteer outreach workers who can screen children for referral to the centre.
• Alerting the team to security risks and changes.

Working with the government

In an acute emergency, it may be necessary to take a more directive role to ensure life saving responses. A commitment to work through the MOH needs to be tempered by a realistic understanding of local ministry capacity. For example, giving full responsibility for the medical components of a TFC to a weak and disorganised MOH will compromise programme quality. Thus, when planning a TFC the following steps should be taken:

• A comprehensive assessment of the MOH's capacity to help, e.g. to receive sick children referred from a SFP, to provide lab facilities for TFC and available nursing and clinical support for the TFC.
• Where MOH skills are lacking, capacity building must be resourced, taking into account staff turnover (which is often high). In the interim, the programme should employ additional medical staff.
• Ensure regular joint management meetings of the TFC with the MOH.
• Ensure that there is clarity regarding any incentive payments for hospital staff.

Monitoring

The impact of feeding programmes on settlement, displacement and migration in pastoral areas should be monitored, and adapted in response to any negative influences. Findings should also be weighed against the programme withdrawal criteria, which in turn should be periodically reviewed and revised as necessary. Regular project review meetings enable the project to adjust to changing situations and context. Clear guidelines are required on how to use monitoring information to manage the performance of feeding programmes. Contextual analysis of the wider political, health, and food security context is the key to responsive programming. Without this, programmes will stagnate or fail to respond to changing conditions.

The contribution of Elizabeth Stevens, Policy Officer at Save the Children UK, in compiling this summary is gratefully acknowledged. For further information, contact Anna Taylor - Nutrition Advisor, Save the Children UK, 17 Grove Lane, Camberwell, London SE5 8RD, UK. Email: a.taylor@scfuk.org.uk

¹Review of Save the Children UK's Emergency Nutrition Programmes in Ethiopia, Kenya and Eritrea, 2000 by Dianne Stevens, August 2001 (then consultant to SC UK, currently SC UK Programme Director in Angola), and South Wollow targeted nutritional support programme, Legambo Woreda: Lessons Identified, October 2000-May 2001 by Kiross Tefera, November 2002 (Emergency Nutrition Officer, SC UK Ethiopia programme)

²Lot quality assurance sampling (LQAS) is a methodology that originated in the manufacturing industry and has been applied to health contexts, such as immunisation coverage. Subpopulations are divided into 'lots' and the sample size is the number of units that are selected from each lot. Before sampling, a decision must be made on the number of "defective" items, e.g. children not immunised, that will deem a 'lot' unacceptable, which in turn will influence sample size. Since the response for each sample is binary, i.e. acceptable or nonacceptable, smaller samples are required compared to other survey methods. By combining information from different 'lots', the LQAS method offers a less conventional means of stratified sampling. (WHO/V&B/0126(2001))