Evaluation of the Supplementary Feeding Programme carried out in Marsabit District, Kenya, April-November 1997

Vincent Brown (Epicentre), Sylvia Carbonell (Médecins Sans Frontières)

MSF implemented a nutrition rehabilitation programme in Marsabit District between April and November 1997, in response to the drought that affected north eastern Kenya in 1996. The primary aim of the programme was to prevent further nutritional deterioration amongst the affected population. The programme catered for over 17,000 children under five years of age and just under 3000 pregnant and lactating women. Children were screened using MUAC measurements. Those with MUACs under 135 mm had weight for height measurements taken and were admitted to the supplementary feeding programme (SFP) if these fell between 70-80% of the reference median (NCHS tables, CDC, Atlanta, USA).

The SFP was designed to account for factors like access to feeding centres and whether beneficiaries were pastoralist or sedentery farmers. Thus, at somefeeding centres the programme was offered as a dry take home ration programme (1,480 kcals per day per child), and at others as either on-site feeding or take home (700 and 1400 kcals/child/day respectively) . Children were discharged on attaining a weight for height of over 85% weight for height on two consecutive fortnightly weighings. The programme was implemented at health centres by MOH or NGO staff and was evaluated mid-term by Epicentre in August 1997.

Methodology

Programme data were analysed monthly - from April to July 1997, in ten areas of the intervention by the evaluation team. As it was not possible to conduct a cluster sample survey due to security/logistic factors the prevalence of global acute wasting was estimated by screening all children in the health facilities of the ten areas. A total of 7,082 children were screened. Children with MUAC measurements below 125 mm were identified as wasted while those under 110 mm were classified as severely wasted.

Results

During this period, monthly prevalence of malnutrition decreased from 15.7% (408/2,596) to 5.2% (47/898) and severe malnutrition from 1.5% (40/2,596) to 0.2% (2/898). Programme coverage was measured at 72% for the district. Between May and July 56.3% of children were discharged while 40.6% defaulted. Deaths and transfers were recorded at 0.4% and 2.7% respectively.
At the end of May a total of 1400 children were registered while the number had dropped to 365 at the end of October. Average duration of stay was 67 days in the take home centres and 71 days in the wet/dry feeding centres.

Discussion

The Drought Preparedness, Intervention and Recovery Programme (DPIRP) had produced data showing the severity of the situation in north eastern Kenya. MSF's decision to intervene was based partly on these findings. The DPIRP collect several types of data including MUAC measurements, livestock growth rates, cereal/meat price ratios and milk intake. The results MSF obtained for MUAC measurements were similar to those obtained by DPIRP, even though the latter measured children between 12-59 months, while MSF measured children from 6 months onwards. The evaluation report recognised that some authors have argued that use of a single cut off point for MUAC will lead to an over-estimation of prevalence of wasting in younger age groups.

The evaluation found that MUAC screening proved an easy way to follow nutritional trends, and was the preferred method due to problems in carrying out a random weight for height survey. It was further contended that this type of screening assessment was particularly applicable to the situation as the population were so dispersed. Furthermore, screening was considered as more or less reliable as 33% of children were checked. Also the decreasing trend in malnutrition was consistent with the decreasing trend in admissions at the feeding centres.

The report pointed out that, although levels of malnutrition rapidly declined in the district there were marked differences between malnutrition rates in different areas. The evaluators also drew attention to the high proportion of defaulters which in their view could be explained by displacements of pastoral groups in search of better grazing areas. Another type of feeding, 'blanket feeding', was implemented for these groups.

Conclusion

By November 1997 the nutrition programme in Marsabit district was gradually coming to an end. The evaluation concluded that the overall programme objectives had been met in a short space of time - seven months, without causing settlement of nomadic groups. Also, as food scarcity and drought is a chronic problem in this area the experience gained with field partners has been very useful for future interventions.

However, authors cautioned, that because of imprecise food relief data it has been difficult to evaluate fully the SFP impact, and that for the pastoralist groups the increase in livestock production registered in recent weeks would also have played a key role in improving the nutritional situation.

Addendum (05/1998) : From November 1997 to April 1998, north eastern Kenya has been badly hit by Rift Valley Fever and malaria epidemics ; this was aggravated by severe climate disturbances (El Nino). As a result, some population groups were again threatened by precarious food security and were severely affected by acute malnutrition. For these groups, Médecins Sans Frontières and other NGOs are currently providing targeted nutrition assistance.

For further information contact: Epicentre, 8 rue St Sabin, 75011, Paris, France. Tel: 40 21 28 50, Fax : 140212803, E-Mail; Vincent-Brown@Epicentre.msf.org