We’re building a new UNICEF.org.
As we swap out old for new, pages will be in transition. Thanks for your patience – please keep coming back to see the improvements.

Evaluation database

Evaluation report

2013 Ethiopia: Assessing the impact on child nutrition of the Ethiopia Community-based Nutrition Program

Author: Jessica White, MPH, and John Mason, PhD

Executive summary

With the aim to continuously improve transparency and use of evaluation, UNICEF Evaluation Office manages the "Global Evaluation Reports Oversight System". Within this system, an external independent company reviews and rates all evaluation reports. Please ensure that you check the quality of this evaluation report, whether it is “Outstanding, Best Practice”, “Highly Satisfactory”, “Mostly Satisfactory” or “Unsatisfactory” before using it. You will find the link to the quality rating below, labelled as ‘Part 2’ of the report.


This report is based on analysis of data collected under UNICEF supervision intended to assess the effectiveness of the Community-based Nutrition component of the Ethiopian National Nutrition Program. The results will be more easily understood in the context of the program itself and the opportunities and constraints of evaluation, in both the Ethiopian and the global context. The report therefore begins with this background, in four sections:
A. A brief description of the Community-based Nutrition program (CBN), and aims of the evaluation
B. Design of the evaluation
C. A summary of the CBN routine data results
D. The global context of nutrition program experience and available evaluations.


The Government of Ethiopia (GoE) legislated a National Nutrition Policy in 2008. This centrally included a National Nutrition Program (NNP), for which by that time there was agreement in principle from the World Bank to provide an initial $30 million in funding over five years (July 2008 – June 20113), expected to be seed money for a much larger investment from other donors. The NNP plan – which had been prepared with World Bank and UNICEF input –laid out a program that was ambitious in terms of implementation (FMOH, 2008). The community-based component was linked to the much larger Health Extension Program (HEP), which trained some 30,000 new Health Extension Workers (HEWs) over the next few years. This was planned to cover most of the country’s 600 woredas (districts) over five years. This in turn was a continuation and evolution of UNICEF and WFP’s Enhanced Outreach Strategy (EOS) which provided a bi-annual campaign to deliver Vitamin A and deworming to children under-5 as well as screening for moderately malnourished pregnant and lactating women (PLW) and children under-5 for referral into the Targeted Supplementary Feeding program (TSF).

The objective of the evaluation research and analysis reported here is primarily to assess the effect attributable to CBN project activities, on outcome indicators, mainly anthropometric. In other words, the impact on young child nutritional status, measured by growth. Effects likely to be associated with the project, on infant and young child feeding (IYCF), maternal health, and certain process indicators have also been analyzed, but not (yet) further investigated in terms of their actual plausible causal attribution (this is for lack of time and resources for continuing analysis). These too are reported.
The estimated changes in indicators are discussed further in relation to the objective indicators in the plans (PAD and PIM), and to expected changes from other data from Ethiopia (e.g. DHS), and other countries.


Following the launching of the CBN in tranches 2 and 3 in 2009-10, four evaluation sample surveys were carried out, two at the baselines and two at midline in September 2011. The four surveys covered about 60 randomly selected clusters each, re-sampling households from the same clusters (referred to throughout as EAs, as they were defined as census enumeration areas) at midline. Thus each EA acted as its own control, and the main outcome estimates were from changes within each EA. No comparison external groups were possible at baseline, according to the government’s policy that no surveys could be done without programs immediately following. However de facto internal comparison groups were possible, from varying implementation. Changes found could also be compared to long-term trends. A number of potential confounders were estimated: economic (as assets), education, environmental, access to services, food insecurity (‘hunger scale’), and others. Regression analysis (OLS) was used to control for the possible effects of these. Data were also available from routine program reporting from weighings; these are used here for guidance, but reported in detail elsewhere.

Findings and Conclusions:

The CBN activities plausibly had a significant benefit on child anthropometry and other aims, well exceeding targets for example for stunting. The decisions that could be based on this are, for the Ethiopian NNP itself, to continue and expand activities, while correcting some weaknesses. The most important is the relatively low participation, around 30%, of children in woredas covered by the program. For policy decisions in Ethiopia and elsewhere, these results join and are supported by findings of other evaluations, gathering increasing evidence that community-based activities aimed at improving child nutrition can be effective. This experience in Ethiopia may be showing a way ahead for many communities in Africa and this needs policy support.


CBN program appears worth continuing, and expanding in area coverage. The data for the first 2-3 years, from evaluation surveys plus CBN routine data, showed that the program exceeded the stated aims, both in terms of implementation and outcomes. Indicators of infant and child feeding practices, and nutritional status objectives, were met or exceeded. More rapid implementation in woredas after the program launch is indicated to increase coverage.
The participation rate in areas where the program is implemented reaches probably 30-40% of the 0-3 year old children. Increasing this participation provides perhaps the greatest potential for increased impact. However some operational research may be necessary to elucidate reasons for the limited participation, and hence ways to increase it.
Attention is needed to initial and refresher training and support to VCHWs. While this was high in the evaluations, this needs to be sustained to maintain incentives.
However, extensive program revision under way, with the ‘Health Development Army’ replacing VCHWs. As we understand it, some nutrition activities will continue to be undertaken by the frontline workers, although the training in nutrition activities may be much reduced (to half a day, it has been suggested). Clearly it is a policy decision as to how far to continue the nutrition activities, which it can now be argued have demonstrated success. Adequate training and monitoring will be crucial as the shift to the Health Development Army proceeds.

Full report in PDF

PDF files require Acrobat Reader.



Report information




Health - Nutrition

UNICEF, World Bank



New enhanced search