Evaluation of SCUK Emergency Nutrition Intervention in Malawi During 2002-2003

Selling maize husks in Malawi

The 2001 harvest in Malawi was particularly poor and it was recognised by June/July that there would be a substantial maize deficit. Furthermore the grain reserve had been sold off, and the money from this reserve was 'missing'.

SCUK conducted a study using the household economy approach (HEA) in late October 2001 in Mchinji district, which suggested that the population were in serious stress. The assessment and verification revealed that 65% of households were suffering from a serious lack of food. Production was down by 40% and terms of trade indicated that between January and October 2001 the cost of a kilogram of maize had increased by 340%. Nutrition surveys were conducted in Mchinji and Salima districts in December 2001. These indicated a deteriorating nutritional status prior to the official hunger gap period with GAM 11.8% and 9.3% and SAM 3.8% and 4.8% in Mchinji and Salima respectively. The rates of severe acute malnutrition were particularly high.

SCUK General Food Distribution Response:

In November 2001 SCUK started advocating for emergency assistance for Malawi. A proposal was submitted to DFID for a two-month food distribution for Mchinji District and it was eventually funded in February 2002. The initial planned response was a two-month general ration distribution in Mchinji District, which was later changed to a one-month distribution. It was planned to distribute to 60% of the population targeting around 400 villages. On registration of villages and beneficiaries the population increased substantially therefore a decision was made to target 50% of the population instead. Distribution commenced in early March. A total of around 45,000 households received a 50kg bag of maize and around 41,000 children received a 10kg bag of Likuni Phala. Village relief committees were set up and community targeting completed.

Follow up nutritional surveys were conducted in March 2002, which indicated that the nutritional status of the under-five children had deteriorated substantially particularly in Salima: GAM increased from 9.3% to 19% (see table 1 below). Over a ten-week period the malnutrition rate had doubled.

This led to a further two-month food distribution funded by DFID planned for May/June targeting 45% of the population. A blanket supplementary food (Likuni Phala) was included as part of the general ration - increasing the ration size to 20kgs per household assuming that there were at least two individuals either under five or pregnant/lactating women.

Mothers and children awaiting assesment in Malawi

The second distribution was planned for May/June when the main staple had already been harvested. A no cost extension was agreed to this programme by DFID until end of December 2002 as by this time other players were also distributing some food. By June 2002 the JEFAP (Joint Emergency Food Aid Programme) had also kicked in with distributions. JEFAP consisted of a consortium of NGOs with Care International the lead agency, and the UN (mainly WFP) importing and distributing food countrywide.

The nutritional status was improving with GAM in Salima down from 19% in March 2002 to 8.1% in June 2002, and in Mchinji it had reduced from 12.5% in March to 5.6% in June. There was a further significant reduction in malnutrition rates in September 2002.

Results and conclusions of the General Ration Distribution:

• Although the general ration in Mchinji was delayed from January to March 2002 it appeared that there was a substantial improvement in the nutritional status. The general ration did not contribute to this as the survey had been completed prior to the food distributions.
• The inclusion of Likuni Phala as a blanket sup plementary food targeted to over 40,000 bene ficiaries was an appropriate choice given the lack of technical and logistical capacity within the MOHP and SCUK to implement targeted supplementary feeding quickly.
• At country level weight for age, a tool for measuring chronic malnutrition was used by the MOHP. It would have been very difficult to change a whole approach to identifying malnutrition countrywide quickly and time constraints were a serious limiting factor.
• Community Managed Targeting and Distribution (CMTD) appeared to work quite well and although time consuming it is a fairer way to target.
• The second food distribution was less timely as it occurred during harvesting but as 65% of the population are poor/very poor it probably supported food security, protected livelihoods and helped to further reduce malnutrition levels.
• Support from the SCUK Regional office in SA was instrumental in procurement of food and logistics.
• A supplies manager was recruited at the start of the emergency who was instrumental in developing plans, forms, etc, making the operation run more smoothly.

Selective Feeding Programmes:

The treatment of moderate and severe malnutrition with targeted feeding programmes was far less successful. Historically, Malawi treats malnutrition in Nutrition Rehabilitation Units (NRUs) attached to health centres and hospitals. The tool for measuring malnutrition in Malawi has historically been weight/age - a measure of chronic malnutrition which is best addressed through longer-term food security/community based interventions. Therefore success rate in these NRUs was pretty awful with high defaulter rates and no proper guidelines. The food pipeline was not always reliable. A similar situation prevailed for treatment of moderate malnutrition conducted at MCH clinic level.

There was little capacity within the NGO or UN communities to address these issues. In early 2002 nutrition task force meetings were instigated. It was recognised that guidelines were fundamental to improving the treatment of acute malnutrition. By August 2002 Malawi draft guidelines for the treatment of moderate and severe acute malnutrition were developed, with support from two external consultants, funded by Unicef. This was a major accomplishment.

SCUK feeding programmes

The records for the supplementary and therapeutic feeding programmes are extremely poor; therefore it was very difficult to analyse coverage of the programme and its impact on reducing malnutrition. It is also difficult to try to understand why the programme was so slow in starting up. From one report, (nutritionist hand-over notes August 2002) it appears that the SFP started in May-June 2002. This is after the hunger gap and not timely. By August SFPs were functioning in 9 centres in Salima and 4 in Mchinji. Data on admissions and discharges and numbers in the different centres in Salima from September 2002 to January 2003 were poor. The figures do not tally indicating poor data collection. The records were substantially better in Salima than in Mchinji.

There are some better statistics from May to August 2003 in both Salima and Mchinji. From the Co-Guard report it appeared that coverage in Salima district was 20-32%, cure rates less than 20% and default rates 70-80%. In Mchinji coverage was higher at around 60%, cure rates around 55% and default rates approximately 30%. These are well below Sphere minimum standards.

With the limited data it is not possible to know overall recovery, length of stay or what weight gains were achieved. It appears that SCUK stopped the SFP in Jan/Feb 2003, as records are only available again in May 2003, under Co- Guard. It is difficult to understand why the SFP was stopped or not functioning properly during this period as this was the 'hunger gap' and there appeared to be funding through the DEC.

Staffing levels was most likely the main constraint. Although there was an international nutritionist during most of the programme period, this person supported two districts and was also involved in national guidelines, technical meetings etc. National technical staff were not recruited until late 2002 up to May 2003. Once these staff were recruited record keeping and reporting certainly improved.

In August/September 2002 a consortium was established with 10 NGO's to co-ordinate and support supplementary feeding programmes. There was an extension to Co-Guard in the form of Co-Guard Two for a further period. However SCUK pulled out of the consortium on the main grounds that the SFP focus was changing to a community based approach, which would be quite different to what was being implemented under the Malawi Nutrition Guidelines. SCUK used some of its final funding from the DEC to support SFP during the hunger gap Jan-March 2004. This was done through the provision of food to the clinics and distributions.

Results and Conclusions of Supplementary Feeding Programme:

• It is extremely difficult to evaluate the SFP intervention, as there is a complete lack of con crete data.
• A main problem was lack of recruitment of qualified staff while lack of other resources such as vehicles and equipment may also have been a problem. It is also possible that the gen eral food aid programme took priority.
• Management was weak.
• Training in the new guidelines, although in draft form in August 2002, did not really take place until May/June the following year when nutrition staff were recruited.
• Visits to supplementary feeding centres at the end of the programme showed that there was general adherence to guidelines. The SFP was extended so that staff trained in the guidelines and data collection would continue and gain experience in this new approach. Although not a particularly logical reason for extending an emergency programme it appeared to build capacity and will have a positive impact on future emergency responses.
• In general there was a lack of planning and exit strategy.

Therapeutic Feeding Programmes:

SCUK provided technical and resource support to the NRUs. Again there are little data on the therapeutic feeding programme apart from records of numbers of admissions and discharges over the period of 2002-2003 in a number of NRUs (Mchinji at Kapiri Hospital, Kochilia and Mchinji hospitals). These three hospitals were visited during this evaluation and staff involved in treatment of malnutrition interviewed. There was adherence to guidelines, drug protocols and feeding schedules. The Multi charts for documenting daily activities were also being used. Staff were very positive about the new guidelines. They felt that cure rates and recovery time had improved substantially. In general, children were in phase 1 for 3-4 days. In Kapiri it was felt that the burden of HIV/AIDs was affecting treatment. With increasing numbers of children HIV positive, more expensive antibiotics were required to treat infections. In all cases eight feeds of milk were given in phase 1, and there was general consensus that the F75 and F100 milks had made it much easier to treat malnutrition. It was much easier to make up feeds as there was no need to cook the milk. This was particularly useful overnight. The results were similar when Mua and Salima Hospitals were visited in Salima district.

Results and comments on Treatment of Severe Malnutrition:

• In general there were very positive outcome for the treatment of severe malnutrition, with adherence to appropriate admission criteria, medical and feeding regimes. This was an enormous achievement.
• Staffing levels had improved in some of the NRUs/paediatric units and the treatment of severe malnutrition has been highlighted.
• Although all medical staff were not trained and there was still some divide between the nutrition and medical rehabilitation, the divide had substantially lessened.
• The Malawi draft guidelines on treatment of severe malnutrition needs to be incorporated into training at nursing and medical schools.
• In general all the NRUs and paediatric units in hospitals supported by SCUK were working well, some better than others. Space is often the biggest constraint particularly during the 'hunger gap' when numbers of admissions increase substantially particularly where the NRU is part of the paediatric ward.
• The intervention has been integrated within the MOHP and is therefore more sustainable although resource support in the form of drugs and the specialised milk may be an issue in the future.

Summary conclusions:

• Over the two-year period the guidelines for treatment of severe malnutrition have been implemented in around fifty NRUs and in 2003 approximately 10,000 children were treated for severe malnutrition. Although the standards do not quite reach those of SPHERE there has been a vast improvement in cure and mortality rates and number of defaulters. SCUK has been one of the players involved in these nutrition gains.
• SCUK were successful in the general ration distribution and were ahead of most other implementing partners in advocating for, and implementing this intervention. However the supplementary feeding programme was far less successful. This may have been mainly due to not gearing up adequately with technical staff.

For further information contact; Frances Mason at; F.Mason@savethechildren.org.uk