CMAM rollout in Ethiopia: the ‘way in’ to scale up nutrition

By Dr Ferew Lemma, Dr Tewoldeberhan Daniel, Dr Habtamu Fekadu and Emily Mates

Dr Ferew Lemma is Senior Nutrition Advisor to the State Minister (Programs), Federal Ministry of Health, and REACH Facilitator, based in Addis Ababa, Ethiopia.

Dr Tewolde has over nine years of experience in the area of nutrition with particular focus on management of acute malnutrition. He has been a Nutrition Specialist with UNICEF Ethiopia since June 2005, previously working with Save the Children and World Vision. He has taken part in the development of Ethiopian national protocol and training materials for management of severe acute malnutrition and development of national guidelines for HIV and Nutrition.

Dr. Habtamu Fekadu is Chief of Party for ENGINE (integrated nutrition programme), Save the Children US, Ethiopia. He has worked in health, nutrition, and academics in Ethiopia for the last 16 years. His considerable portfolio of experience includes Federal Ministry of Health nutrition lead on the five year National Nutrition Programme (NNP) of Ethiopia, amongst a broad range of other activities including strategy development, training and evaluation, and working with other agencies, notably UNICEF and Save the Children.

Emily Mates is a public health professional with a focus in nutrition. She was lead researcher with ENN on the CMAM Conference based in Addis Ababa, where she has worked for many years in emergency and development health and nutrition programming.

The authors would like to mention in particular the support of Dr Abdulaziz and Mesfin Gose (Federal Ministry of Health), Sylvie Chamois (UNICEF), Pankaj Kumar and Israel Hailu (ConcernWorldwide) Iassack Manyama and colleagues (ENCU/ DRMFSS) and the many other partners implementing CMAM in Ethiopia.

Acronyms:
ASRI	Accelerated Stunting Reduction Initiative
CBN	Community Based Nutrition
CHD	Community Health Day
CMAM	Community Management of Acute Malnutrition
EDHS	Ethiopian Demographic and Health Survey
DRMFSS	Disaster Risk Management and Food Security Section
EHNRI	Ethiopian Health and Nutrition Research Institute
ENCU	Emergency Nutrition Coordination Unit
ENA	Essential Nutrition Actions
EOS	Enhanced Outreach Strategy
FFA	Food Fortification Alliance
FMoH	Federal Ministry of Health
GAM	Global Acute Malnutrition
GMP	Growth Monitoring and Promotion
GoE	Government of Ethiopia
HEP	Health Extension Programme
HEW	Health Extension Worker
ICCM	Integrated Community Case Management
IMNCI	Integrated Management of Neonatal and Childhood Illnesses
INGO	International NGO
IRT	Integrated Refresher Training
IYCN	Infant and Young Child Nutrition
LBW	Low Birth Weight
MAM	Moderate Acute Malnutrition
MDG	Millennium Development Goal
MOH	Minstry of Health
MUAC	Mid Upper Arm Circumference
NCHS	National Centre for Health Statistics
NGO	Non-Governmental Organisation
NNP	National Nutrition Programme
NNS	National Nutrition Strategy
OTP	Outpatient Therapeutic Programme
RHB	Regional Health Bureau
RUTF	Ready-to Use Therapeutic Food
SAM	Severe Acute Malnutrition
TFP	Therapeutic Feeding Programme
UNICEF	United Nations Children’s Fund
VAS	Vitamin A Supplementation
WFP	World Food Programme
WHO	World Health Organisation
WoHO	Woreda Health Office
ZHD	Zonal Health Department

Introduction

Globally, more than 3.5 million children under the age of five year die each year due to the underlying causes of malnutrition. It is also estimated that 13 million infants are born each year with low birth weight (LBW), 55 million children are wasted (of which 19 million are severely wasted) and 178 million are stunted. Of the estimated 178 million stunted cases, 90% live in 36 high burden countries that include Ethiopia¹. The consequence of the many adverse interacting elements in Ethiopia is that although malnutrition rates among children are steadily decreasing, they remain at unsatisfactorily high levels. The 2010 Ethiopian Demographic Health Survey (EDHS)² estimated the national prevalence of Global Acute Malnutrition (GAM) at 9.7%, with 44.4% of children estimated to be stunted and 28.7% underweight. Encouragingly, both underweight and stunting prevalence was reducing by 1.34% per year over the past decade. While this trend is clearly progressing in the right direction, Ethiopia will only reach the Millennium Development Goal (MDG) target of halving the number of underweight children if the percentage reduction is increased to at least 1.6 percentage points per year. This implies the need to intensify and scale-up known high impact nutrition interventions and those that address wasting. Figure 1 describes the changes in malnutrition prevalence from 2000-2010.

* Recalculated using World Health Organisation (WHO) Growth Standards⁴ for 2000 and 2005

The Government of Ethiopia (GoE) has developed a five-year development plan, the Growth and Transformation Plan (GTP), for the period 2010/11 to 2014/15. The main objectives of the GTP include:

1. Maintain an average real Gross Domestic Product (GDP) growth rate of 11% and attain the MDGs
2. Expand and ensure the quality of education and health services and achieve MDGs in the social sector
3. Establish suitable conditions for sustainable nation building, through the creation of a stable democratic and developmental state
4. Ensure the sustainability of growth by realizing all of the above objectives within a stable macro-economic framework.

Under the umbrella of the GTP, the GoE launched the fourth Health Sector Development Programme (HSDP-IV). The new (and final) HSDP IV (2010 - 2015) places a strong focus on maternal health issues and has considerably more focus on nutrition than the three previous plans. There are 16 nutrition indicators within HSDP-IV, examples of which include reducing the stunting prevalence from 46% to 37%, reducing the prevalence of wasting from 11% to 3%, and increasing household utilisation of iodised salt from 4% to 95%³.

During the course of implementation of the previous health sector development programme (HSDP-III 2005/6 - 2009/10), a National Nutrition Strategy (NNS) was developed and launched in 2008. The NNS is operationalised through the National Nutrition Programme (NNP), a 10- year initiative aiming to reduce the levels of stunting, wasting, underweight and LBW infants. The first phase is for five years (2008-2013), at an estimated cost of 370 million USD and consists of two main components: ‘Supporting Service Delivery’ and ‘Institutional Strengthening and Capacity Building’. The overall objective is better harmonisation and coordination of the various approaches to manage and prevent malnutrition.

The service delivery arm of the NNP has four sub-components: a) Sustaining Enhanced Outreach Strategy (EOS) with Targeted Supplementary Food (TSF) and transitioning of EOS into the Health Extension Package (HEP), b) Health Facility Nutrition Services, c) Community Based Nutrition (CBN) and d) Micronutrient Interventions.

A process of revision and extension of the NNP has recently commenced (October 2011) for two main reasons:

1. To align the end of the first phase with the HSDP IV and MDGs, i.e. extend the first phase by 2 years to 2015
2. To strengthen initiatives that were not adequately addressed in the original document and include initiatives that have emerged since the NNP was devised. For example:

• Accelerated Stunting Reduction Initiative (ASRI) - inclusive of maternal nutrition, Infant and Young Child Nutrition (IYCN)
• Food Fortification Alliance (FFA), goals and objectives for improving micronutrient status
• Strengthening of multi-sectoral linkages - key sectors include; agriculture, education, water and energy, labour and social protection, finance and economic development, women’s children and youth affairs
• Social protection policy and nutrition related indicators
• Moderate acute malnutrition (MAM) programming and the development of improved linkages between preventive and treatment programming
• School health and nutrition (SHN)

CMAM/TFP roll-out in Ethiopia

The term Therapeutic Feeding Programme (TFP) is used in Ethiopia to describe the treatment of Severe Acute Malnutrition (SAM). Much has already been written about Ethiopia’s scale up experience to date⁵, so the history and development of the TFP in Ethiopia is only briefly summarised here. Community based management of acute malnutrition (CMAM) in Ethiopia traditionally does not include the management of MAM. Hence the discussion below focuses on SAM management only.

A small pilot for CMAM was first conducted in Southern Ethiopia in 2000. A research programme in three countries (Malawi, Ethiopia and South Sudan) followed, implemented from 2002 by Valid International and Concern Worldwide, to test the efficacy and safety of the CMAM approach.

A food security crisis due to drought developed across many areas of the country during 2003/4. This crisis was the catalyst for many international non-governmental organisations (INGOs) to adopt the CMAM approach of treating the majority of cases as outpatients, as they became overwhelmed trying to manage the high caseloads of malnourished children arriving at the Therapeutic Feeding Centres (TFCs).⁶

From 2004/5, the Federal Ministry of Health (FMoH), alongside partners including UNICEF and others, commenced scale-up of SAM treatment services. This involved developing guidelines and establishing more in-patient and out-patient services across the country. In 2007, following international endorsement of the CMAM approach,⁷ the national protocol for SAM treatment was revised to include detailed guidance for the Outpatient Therapeutic Programme (OTP) and community mobilisation activities.

In 2008, a dramatic and rapid increase of SAM cases was seen across Oromia and Southern Nations, Nationalities and People’s (SNNP) regions as food security deteriorated due to drought. Responding to this emergency by maximising access and coverage of these lifesaving services, the FMoH reviewed the evidence of CMAM effectiveness when implemented at health centre level and made the decision to decentralise CMAM services to primary health care (health post) level. This involved OTP managed by the Health Extension Workers (HEWs)⁸, as outlined in Box 1.

 

To achieve the rapid decentralisation of OTP, the FMoH led the development of simplified quick reference materials in July 2008⁹. This was immediately followed by a national level master training for nutritionists from NGOs and Regional Health Bureaus (RHBs) to enable cascading of training in 100 districts (woredas) in Oromia and SNNP regions. The master trainers facilitated regional level Training of Trainers (ToT) sessions. The trained staff then provided two-day training for district and HEW staff. By November 2008, 455 health posts in the two affected regions were managing OTP, with results reaching International Sphere recommendations for selective feeding programmes.¹⁰ These good results prompted major and accelerated efforts for scale-up of the TFP across the other two main regions (Amhara and Tigray). The pace of scale up has continued, with > 8,000 health facilities currently offering OTP services across Ethiopia. Table 1 shows the number and coverage of health facilities providing CMAM services in Ethiopia.

Table 1: District level coverage of TFP/CMAM in Ethiopia, October 2011
Hotspot priority number11	Number of districts	Number of OTPs	Number of SC/TFU
1	175	3,106	192
2	138	2,677	147
3	40	655	32
4	269	1,662	102
TOTAL	622	8,100	473

 

The FMoH has guided the roll-out of the TFP. It is no longer viewed as a response necessary in times of emergency only. Instead it has become part of the integrated national approach of decentralising primary health care services across the country, through the Health Extension Programme (HEP). This is described further below.

Results of national TFP scale-up

A total of 731,238 severely malnourished children were admitted to the TFP between January 2008 and September 2011, as outlined in Figure 2.

Figure 2 clearly illustrates that the number of children admitted each month continued to increase with the increasing number of OTP sites, while at the same time showing the seasonal variation of caseloads in Ethiopia.

The performance of the TFP has been highly successful with impressive programme results: an average recovery (cure) rate of 82.3%, mortality rate of 0.7% and defaulter rate of 5.0%. All results are well above the Sphere international recommendations, a major achievement for this government-led national public health initiative.

Consistently low mortality rates provide evidence of the ability of primary health care workers to identify and refer sick children - those with a lack of appetite or additional medical complications that require higher-level health care. Note that the low mortality rate is also related to the early case detection that comes from having massively decentralised services. Caregivers can access assistance earlyon in the disease process of their child, reducing the need for referrals of complicated cases for in-patient care as well as the risk of death.

The low default rates also confirm the reduced opportunity costs for caregivers when services have been decentralised at scale. These low default rates (for a programme that requires more than one visit to the health facility) also demonstrate broad community confidence in the programme.

The wide-scale roll out of TFP/CMAM in Ethiopia allowed for early detection of the deteriorating nutrition situation during the 2011 Horn of Africa crisis, through identification of the rapidly increasing admission trends in SNNPR and Oromia regions. The country was better prepared to mobilize resources and further develop the capacity already built, well before the crisis was declared globally. Most importantly, the efforts made over the past few years to decentralise TFP/CMAM in Ethiopia ensured that many deaths related to SAM during this current crisis have been averted.

An enabling context for the national TFP scale up - The Health Extension Programme

HSDP III has been a triumph for primary health care in Ethiopia, with massive roll-out of the Health Extension Package (HEP). The HEP involved the training and deploying of 33,000 female HEWs to strengthen the primary health system (1 HEW per 2,500 population, 2 HEWs working together at each village health post). The HEP is well-established across the country and some evidence of its success can be seen in the preliminary results of the EDHS 2010, showing a sustained decrease in infant and under-five mortality rates.¹²

The HEP was originally designed for preventative activities only. The health leadership in Ethiopia has proven to be adaptable when presented with solid evidence, e.g. TFP/CMAM programming (that was decentralised to health post level from 2008) and early treatment of diarrhoea, malaria and Acute Respiratory Infections (ARI). The role of the HEWs has now been formally widened to include basic treatment services as outlined in the Integrated Community Case Management (ICCM), which has been included in the Integrated Refresher Training (IRT) package currently being delivered in a phased approach to HEWs across the country. This heralds the full integration of TFP/CMAM into the public health system in Ethiopia where a severely malnourished child can access treatment in any health facility in the same way as a child with malaria.

The TFP reporting system

The rapid expansion of the TFP (from 1,240 sites at the end of 2008 to 4,325 by the end of 2009, a 240% increase) ensured that the focus needed to remain on training and capacity building of HEWs and supervisory staff in managing SAM treatment at health post level. Partners were well aware that the reporting system (designed to monitor the number of sites implementing the programme and the quality of care, through tracking recovery, death, default) was poorly functioning during the first two years, but the focus was necessarily on the capacity building of health staff. At the beginning of 2010, as the numbers of TFP sites continued to expand, it became a priority to improve the reporting rate.

UNICEF recruited a TFP Reporting Officer for each region (initially for three months but extended to 11 months of 2010), operating under the Emergency Nutrition Coordination Unit (ENCU). The reporting rates significantly improved, in part due to the TFP Reporting Officers who worked to identify the bottlenecks in the reporting system. In the short term, they also acted as ‘couriers’ for the data early in 2010. See Figure 3 for the progression of TFP expansion and reporting rate.

In order to sustain this improved reporting rate from the regions, the ENCU conducted a review in 2011 to document the lessons learned of how the TFP reporting rate improved. Some of the key lessons included the need for:

• Continuous advocacy on the importance of timely and accurate TFP reports at regional and woreda levels, by all nutrition staff in the regions.
• Training of zonal and woreda Maternal and Child Health (MCH) experts in use of the TFP data base and completion of monthly reports and providing supportive supervision for relevant staff.
• Including reporting rates as one of the performance evaluation indicators amongst health workers.
• Discussion of reporting rates in the monthly and quarterly review meetings held at regional level, including analysis of reporting submission to encourage the close follow up for those facilities/woredas not reporting.
• Continuous follow up and regular communications with woredas an health facility level experts, using all available means (telephone, e-mail, fax and other networks).

The benefits of the efforts towards improving the reporting rate (consistently above 80%), is that there is now trend data which shows the impressive expansion and successful performance of the TFP at primary health care level.¹³ Additionally, widespread coverage and accurate reporting of the TFP is providing invaluable trend monitoring data. In the absence of routine nutrition information (see below, challenges) reports of increasing numbers of admissions to the TFP have become crucial data alerts for authorities to deteriorating situations, as seen in the lowland drought affected areas during 2011.

There remain on-going challenges for the TFP reporting and nutrition information systems. Although the reporting rates have remained consistently above 80%, there is often a delay in timely compilation and submission of reports. The information often comes late, reducing its efficacy for ‘early warning’ of deteriorating situations. Also, the standardised database for TFP monitoring is only at regional level and has not yet been implemented at woreda level. With the expanding numbers of TFP sites, there is increased importance for this trend monitoring data to be accurate and timely.

There are also opportunities for the reporting systems. The HMIS has been revised and now includes TFP data in a manner that enables tracking performance standards against the Sphere indicators. Moreover, HMIS reporting from woreda to regional levels will soon change from a quarterly to monthly basis. This will create a solid opportunity to fully integrate TFP/CMAM reporting into the national HMIS.

Linkages with other programmes

TFP/CMAM in Ethiopia has developed some linkages with other nutrition programmes that are implemented under the umbrella of the NNP including:

Community Based Nutrition (CBN)

CBN is the preventative arm of the nutrition service delivery outlined in the NNP. It aims to use community capacity to assess and analyse the nutrition situation of its own community and take appropriate action. Monthly Growth Monitoring and Promotion (GMP) sessions, followed by community conversations and counselling, are used as tools to elicit the triple-A cycle of assessment, analysis and action. The programme has been gradually expanded, training over 90,000 Community Health Volunteers (CHVs). CBN has been scaled-up to 228 woredas in the four main regions of Ethiopia (SNNP, Tigray, Amhara, Oromiya) supported by development partners of the FMoH.¹⁴ In 2012, the CBN will be rolled-out to an additional 115 woredas bringing the total number of woredas to 343. UNICEF provides technical assistance and support for government implementation.

OTP training is provided as part of CBN training in the 343 CBN woredas where CBN is implemented, creating an opportunity for both programmes to benefit from this linkage. The community conversations within the CBN are proving useful in assessing and analysing why a child is malnourished and what behavioural changes could foster improved nutritional status for the children in a family, using their existing resources. Additionally, the presence of TFP/CMAM in all CBN woredas provides good opportunities for referrals and behavioural change messaging for severely malnourished children.

Enhanced Outreach Strategy (EOS)

The Enhanced Outreach Strategy/Targeted Supplementary Food Programme (EOS/TSF) was designed and initiated jointly by the FMoH, the Disaster Risk Management Food Security Sector (DRMFSS) (former Disaster Prevention and Preparedness Agency), UNICEF and WFP, to address some of the most critical child survival and malnutrition problems in Ethiopia and to provide a bridge to sustained nutrition interventions through the HEP. The EOS was launched in April 2004 with the aim of reducing mortality and morbidity in 6.8 million children under 5 years, as well as pregnant and lactating mothers in 325 drought prone woredas across Ethiopia. This was to be achieved through the implementation of key child survival initiatives, including Vitamin A Supplementation (VAS), de-worming, measles vaccination and screening for malnutrition and subsequent treatment of malnutrition. A major success of the EOS programme has been Vitamin A coverage consistently recorded as over 80% since 2005.

The EOS has transitioned into Child Health Days (CHD) in the 228 Woredas where the CBN programme is currently being implemented. To facilitate the transition of more EOS woredas into the CHDs, an operational plan for transition has been prepared and is under discussion between the FMoH and key partners.

Using the opportunity presented by the six-monthly VAS campaigns, screening for acute malnutrition using Mid-Upper Arm Circum- ference (MUAC) in drought- affected woredas is also undertaken. Children and pregnant and lactating women (PLW) identified as moderately malnourished receive 3-monthly supplementary food rations through the TSF, while those identified as severely malnourished are referred to the nearest health facility providing TFP/CMAM services. The number of woredas implementing the TSF component of the EOS has been reduced to 167 drought affected woredas in six regions. This is largely due to the lack of sufficient resources available to procure and supply supplementary rations. A concept note has been developed by the FMoH, DRMFSS, UNICEF and WFP regarding the transition of TSF into a programme for management of MAM in the medium to long term.

RUTF in Ethiopia: supply, importation, local production and distribution mechanisms

The development and use of RUTF has been the critical factor that helped to revolutionise the management of SAM, through enabling outpatient treatment for the vast majority of malnourished children. From 2003 to 2005, INGOs generally provided their own supplies for the projects they implemented.

By 2005, the OTP was slowly being scaled up. During the hunger gap in the same year, UNICEF was required to air-lift approximately 400 metric tons of RUTF from their European supplier. In addition to the extra costs associated with air-freight, complicated and time-consuming customs clearance processes presented a challenge for the importation of RUTF. UNICEF took on the role of central procurer and distributor of RUTF for most organisations to facilitate the importation processes. UNICEF procured and distributed a total of 11,472 metric tons of RUTF between January 2008 and September 2011¹⁵.

Small scale local production of RUTF was piloted from 2004/5 by Concern and Valid Nutrition, using a small scale local producer and locally produced raw materials, except for the Dried Skimmed Milk (DSM) and mineral/vitamin mix which had to be imported. However, these pilots were unsuccessful as it proved difficult to ensure the quality of the product using small-scale producers.

Success factors for local production

In early 2007, larger-scale production was established following an initial investment from a US-based philanthropist (donating over 300,000 USD, to be repaid back to UNICEF through in-kind contribution by the local manufacturer once the production was up and running). Through the use of Nutriset’s franchise network (plumpyfield), a local company HILINA received the transfer of technology and skills from Nutriset that enabled local production of RUTF, of a quality that passed the expected standards of both Médecins Sans Frontières (MSF) and UNICEF.

While the local producer was gradually scaling-up production, it was not enough to meet the needs of the expanded TFP during nutrition stress years (such as 2008). A large amount of RUTF still needed to be imported, although the proportion supplied by local production is encouraging.

Between January 2008 and June 2011, approximately 39.3 million USD had been invested in the procurement of RUTF. This cost does not include the freight and distribution expenses. RUTF remains the most expensive component of the TFP; a cost analysis is currently being undertaken (together with the CMAM evaluation), which is expected to provide more information of the costing associated with the TFP in Ethiopia.

Challenges with local production

The local producer continues to procure all peanuts and oil from the local market, which positively contributes to the local economy and livelihoods of farmers. However, sometimes the quality of the RUTF has been compromised, with unacceptably high levels of aflatoxin contamination from poor handling and storage of peanuts. The local producer has taken several steps to ensure that levels of aflatoxin stay within acceptable recommendations. UNICEF has also instituted a system of testing each and every batch of RUTF for contamination. This has resulted in a two week lead time after completion of the production until aflatoxin test results are received from an independent laboratory in the UK. These efforts by the producer to improve the quality of the locally sourced raw materials have been showing results. Over the past 12 months, only one batch of RUTF has failed to comply with acceptable levels of aflatoxin in the final product.

Distribution systems and structures for RUTF

The in-country distribution of RUTF uses various routes to reach the health facilities. The bulky nature of the RUTF in both volume and weight that is required to ‘cure’ each severely malnourished child is considerably larger than the drug supplies usually needed for routine treatment of other lifethreatening conditions. As a result, pre-positioning several months worth of RUTF supplies has often been beyond the warehousing capacity of the health system. Additionally, the seasonal and sometimes drought-related rapid increases in admissions to the TFP, intensifies the pressure on the health service logistic system for ensuring timely deliveries of large volumes of RUTF.

The FMoH uses the Pharmaceutical Fund and Supplies Agency (PFSA) logistic system for most medicines and supplies used within the health system. As described, RUTF is a bulky and heavy product, which has meant that it is beyond the current capacity of the PFSA system to handle distribution and storage. As a result, UNICEF and partners have been required to deliver the RUTF through the RHBs and ZHDs, indicated in Figure 6.

UNICEF has distributed an average of approximately 2,800 metric tons of RUTF per year since 2008 to health facilities across Ethiopia.¹⁶ Mostly it is delivered directly to the RHB warehouses although in times of emergency, UNICEF sometimes delivers to the zonal level or direct to woredas (dotted lines in Figure 6), to minimise the risk of damaging stock-outs. Re-supplying of the RUTF is based on official requests from the RHBs using the TFP reporting system, with re-supply levels based on the monthly reported caseloads.

Major successes of the RUTF supply and distribution system

The system has enabled rapid expansion of CMAM capacity to over 7,000 health posts. It is flexible and able to respond to emergency needs. Performance is strongly related to the technical persons implementing the programme, as they take the lead in requisitioning and distributing the RUTF. NGOs can access the RUTF from ZHDs or RHBs and support its delivery to health post level. UNICEF acting as the central procurement channel has considerably eased the burden on partners for importation and customs clearance.

Major challenges of the RUTF supply and distribution system

The limited warehousing capacity of the regional and zonal health offices can sometimes affect the quantity of RUTF that can be delivered and stored safely. Late requests and inadequate forecasting of projected consumption compromise programming. Some misuse/ leakage of RUTF by clients has been reported (selling and sharing), using for moderately malnourished children and at times, adults. Some duplication can occur between partners, e.g. UNICEF, the Food By Prescription programme (FBP) and GOAL, creating difficulties for some facilities to track records of clients versus commodities. Coordination meetings have been established to assist with reducing duplication.

CMAM transition in emergencies and development

Management of SAM has traditionally been considered an emergency response, often implemented by NGOs. In the context of chronic food insecurity and seasonal hunger, programmes open based on emergency thresholds of SAM and GAM rates and then close as the situation improves, only to reopen in the next hunger season. The implications of this traditional emergency focus of CMAM include irregular and short-lived funding, inadequate resources for capacity building of the health system and delays in the emergency response. These delays have mostly been linked with the time needed to identify the affected woredas and conduct nutrition surveys, in order to justify the poor situation and hence access emergency funding from the various donors. This paradigm has resulted in additional costs of repeatedly phasing in and phasing out of programmes for the management of acute malnutrition in chronically affected woredas. The timeliness and adequacy of RUTF provision can be hostage to the declaration of emergency situations and resulting donor pledges. Hence there is a need for improved funding mechanisms, especially for on-going situations that may not be characterised as a humanitarian emergency.

The extensive CMAM rollout in Ethiopia has enabled unusual access to longitudinal information on admissions of severely malnourished children to public health facilities over the past few years. Instead of waiting for nutrition surveys to be planned, undertaken and compiled, humanitarian actors can easily identify the progress or deterioration of a given nutrition situation, through surveillance of the monthly admissions to CMAM. The massive increase in coverage of CMAM services across the country has allowed access to first-hand information from wide areas. These constitute a considerable proportion of the country, especially if compared to the handful of woredas that were being reached through nutrition surveys. However, it must be noted that routine programming data, reports and anthropometric measurements will likely be of lower quality than standard nutrition survey data. Therefore, while the use of nutrition survey data remains relevant in specific situations, it is not necessarily the only tool available for decision making for action.

In addition to nutrition surveys, hot-spot classification has been introduced in Ethiopia. The ‘hot-spot’ priority list provides the basis for the Relief Requirement Plan released by the DRMFSS in collaboration with all sector ministries and the UN. The use of the ‘hot-spot’ classification system has been a step forward from the sole reliance on the use of GAM and MAM thresholds, to decide when to start and stop interventions.

Ways forward

Integrated management of acute malnutrition at scale

CMAM has integrated very well into the primary health care system of Ethiopia and is undoubtedly saving the lives of many vulnerable children. There has been demonstrated success when linking CMAM with the Integrated Management of Neonatal and Childhood Illnesses (IMNCI) and ICCM initiatives. Encouragingly, many opportunities for the capacity building of frontline health workers continue to present themselves in Ethiopia. What is less clear is how the level of funding for integrated treatment for SAM will be sustained over the longer-term, since the supplies are expensive. There is an urgent need to strategise the possibilities of funding sources beyond humanitarian mechanisms. This could not only provide funding sources for ongoing needs, but would enable more equity of services, if severely malnourished children in ‘non-emergency’ woredas were able to have the same access to treatment as those living in identified hot-spot woredas. The cost analysis of the UNICEF/MOH CMAM evaluation (currently underway) is expected to provide useful insights on the cost effectiveness of investing in the management of severe acute malnutrition.

The implementation of TFP/CMAM at scale calls for concerted efforts and investment in quality monitoring and improvement. CMAM quality improvement is contingent on many of the health system pillars¹⁷ including service delivery, information systems, the health workforce, medical products, health financing and leadership. As a result, efforts to improve CMAM quality should be viewed from the health system’s perspective, and therefore contribute to overall improvements in the system.

In addition, there is a need for improved linkages between TFP/CMAM, CBN and other direct nutrition interventions currently being implemented in Ethiopia to ensure that the maximum gains are being leveraged from the considerable investments being made by both government and partners.

Operational research priorities

Under the NNP, operational research is identified as crucial for developing our understanding of effective preventive and curative nutrition interventions. A number of research possibilities have been identified by FMOH/ EHNRI and partners, with priority operational research areas as follows:

• Cost effectiveness study of TFP/CMAM in Ethiopia
• Determinants of successful and lasting management of SAM through community based nutrition activities
• Assessment of quality of nutrition data; flow, data utilisation, and validation
• Study on the effectiveness, feasibility, acceptability and compliance of micronutrient powders (e.g. Sprinkles) to improve complementary feeding practices and reduce micronutrient deficiencies in children under 2 years of age.

Conclusion

The large numbers of severely malnourished children successfully treated over the last few years testifies to Ethiopia’s success in fully integrating the out-patient management of SAM into all levels of the routine health system. Importantly, across this vast land, services have been decentralised to primary health care level to improve access and coverage.

Based on our successful experience of scaling up TFP/CMAM in Ethiopia, countries that are considering starting TFP/CMAM could try to scale-up services to national level. Such actions save lives, both during emergency situations and as part of routine nutrition interventions. It is clear that the implementation of TFP/CMAM at-scale not only puts pressure on the health system, but also stimulates it to respond to the additional demands. This could be due to the fact that the programme is so visibly successful; it creates demand from within communities because of the rapid improvement in their sick malnourished children; when able to access appropriate treatment, the transition of their children - from listless and lethargic, to playful and energetic - can provide a powerful motivating force for the community.

Ethiopia has learned that to successfully rollout TFP/CMAM, it is vital to ensure government commitment and to develop good coordination between government and development partners (especially for resource allocation). It is also crucial to create a wellestablished logistics system and well thought-out monitoring and evaluation systems, to ensure both quality and continuity of services.

For more information, contact: Dr Ferew Lemma, email: ferew.lemma@yahoo.com

---

¹Black, R, Allen, L. H, Bhutta, Z. ., Caulfield, L. E, De Onis, M, Ezzati, M, Mathers, C, and Rivera, J. For the Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: global and regional exposures and health consequences. The Lancet. Published online Jan 17 2008. DOI:10.1016/S0140- 6736(07)61690-0

²This data is not yet official until the full EDHS report 2010 is issued (expected December 2011).

³Recalculated by Tulane University.

⁴As the DHS 2010 was not out during the HSDP-IV preparation, DHS 2005 was used as a benchmark.

⁵Field Exchange issue 40. Emergency Nutrition Network. http://fex.ennonline.net/40/contents.aspx

⁶TFCs were often established in a health centre compound with erection of a large tent, and heavy presence of NGO staff to manage the cases on a daily basis.

⁷WHO/WFP/UNSCN/UNICEF. Community-Based Management of Severe Acute Malnutrition. A Joint Statement by the World Health Organization, the World Food Programme, the United Nations System Standing Committee on Nutrition and the United Nations Children’s Fund, 2007. http://www.who.int/nutrition/topics/statement_combased_malnutrition/en/

⁸Sylvie Chamois (2009). Decentralisation of out-patient management of severe malnutrition in Ethiopia. Field Exchange, Issue No 36, July 2009. p12. http://fex.ennonline.net/36/decentralisation.aspx

⁹See Footnote 8 for details of the rapid decentralisation process in 2008.

¹⁰Of 27,739 SAM children treated, rates of 77.6% recovery, 0.7% mortality and 4.2% defaulter. The Sphere Project recommends recovery >75%, mortality < 10%, defaulter < 15%, coverage of >50% in rural communities, >70% in urban populations and >90% in a camp situation. The Sphere Project. Humanitarian Charter and Minimum Standards in Disaster Response. Geneva, 2011 Edition. Sphere Project. www.sphereproject.org

¹¹A ‘hot-spot’ classification system has been introduced in Ethiopia where woredas are classified using concepts from the IPC (Integrated Phase Classification) approach. The emergency affected woredas are ranked based on the level of existing hazards including current food security, disease outbreak, flooding, CMAM admissions, nutrition survey results and other related indicators. Emergency affected woredas are classified as priority 1, 2 and 3 woredas, while non-emergency woredas are classified as priority 4.

¹²Since 2005, infant mortality has decreased by 23%, from 77 to 59 deaths per 1,000 live births. Under five mortality has decreased by 28%, from 123 to 88 deaths per 1,000 births. EDHS preliminary results, 2010.

¹³A considerable effort was also placed on establishing a monitoring system for the TFP. Independent field monitoring officers worked alongside RHB and woreda officials using standardised checklists and scorecards. A detailed description of this is provided in Field Exchange issue 40, pages 38-42. See footnote 10 for full reference.

¹⁴Development partners providing support include the World Bank, UNICEF, CIDA, Dutch Government and JICA.

¹⁵See article regarding UNICEF global supply of RUTF including Ethiopia in Field Exchange 42. Increasing access to RUTF. Jan Komrska, UNICEF.p46-47.

¹⁶This is equivalent to over 217,000 cartons or over 32.5 million sachets per year

¹⁷WHO. Everybody’s Business: Strengthening health systems to improve health outcomes: WHO Framework for action. 2007. (accessed at http://www.who.int/healthsystems/round9.2.pdf