Enable low bandwidth mode Disable low bandwidth mode
FEX 40 Banner

Integrating OTP into routine health services CONCERN’s experiences


By Emily Mates

Emily Mates is a public health professional with a focus in nutrition. She recently left Concern Worldwide, Ethiopia where she worked for many years in emergency and development health and nutrition programming

The author would like to thank the Ministry of Health in Ethiopia, who so often showed impressive commitment for a new service that added to their already heavy workload. With thanks also to the staff of CONCERN; a privilege to work with a great team, where no task was too much nor distance too far to travel in the pursuit of supporting quality services for malnutrition.

CONCERN is especially grateful to OFDA for the generous support (both financial and otherwise) that enabled the N-CMAM programme to happen.

This article describes the history of an international non-governmental organisation (INGO)-implemented programme that evolved from an innovative emergency intervention into a longer-term initiative, to support the Ministry of Health (MoH) to integrate treatment services for Severe Acute Malnutrition (SAM) within the routine health system. The article describes the community-based management of acute malnutrition (CMAM) approach, a timeline of CMAM programming in Ethiopia and CONCERN's national CMAM (N-CMAM) programme and a description of the N-CMAM programme goal, activities and programme results. It concludes with a discussion of key factors contributing to the enabling environment and analysis of challenges, lessons learned and programme limitations.

The approach

The N-CMAM programme was implemented using a partnership approach, with the focus firmly on 'ownership' by the MoH. The aim was for the programme to establish a much needed service during 'normal' times and to provide a base of capacity from which services could be rapidly scaled-up at times of crisis. The programme provided a package of 'minimal support' to the MoH, which consisted of training (set-up, on-the-job and training-oftrainers (ToT)), joint supervision, workshops, experience sharing visits and community mobilisation support. The programme emphasised learning and innovation, in order to regularly refine and adapt the approach. The programme also provided considerable support to CONCERN's emergency nutrition interventions.

SNNPR MoH team on experience exchange in Tigray region

While the minimal support package was important, the most crucial aspect of the programme was the nature of the dialogue between the partners. CONCERN attempted to find the right balance between providing the support that was required and ensuring that the MoH were in the 'driving-seat' at all times. In this way a strong partnership developed, with all levels of the MoH steadily gaining confidence in their ability to offer quality CMAM services. When the food crisis of 2008 developed across much of the country, the MoH (with support from UNICEF and partners) were able to rapidly scale-up decentralised CMAM services through the national Therapeutic Feeding Programme (TFP) 'roll-out' and implement the required policy changes. To date, approximately 50% of health facilities are offering CMAM services, a huge achievement across a vast country.

CMAM/TFP in Ethiopia

CMAM programming is the internationally recommended way of treating severe acute malnutrition (SAM).1 In Ethiopia, the national TFP (Therapeutic Feeding Programme) roll-out is being implemented using the CMAMapproach, as such the terms CMAM and TFP are often used interchangeably (see Box 1). CMAM/TFP combines in-patient and outpatient care for children (6-59 months) suffering from SAM (mid upper arm circumference (MUAC) <11.0cms, weight for height (WFH) <70%, and/or bilateral pitting oedema). It is recommended that there is at least one inpatient unit located in each district (woreda). An OTP (Outpatient Therapeutic Programme) site is ideally established in the health posts located in each village/kebele, staffed by community-based Health Extension Workers (HEW). The CMAM approach recommends extensive community mobilisation, as the success of OTP is dependent on a wellinformed and responsive community.

Box 1: Technical description of TFP/CMAM in Ethiopia

Dawit tucking into RUTF during appetite test

The majority of children with SAM who have no medical complications and have a good appetite (around 90%) are treated as outpatients from their local health posts and health centres through the OTP. OTP services are provided on a weekly basis, where the children receive a medical check and rations of Ready to Use Therapeutic Food (RUTF), most commonly the product Plumpy'nut®, according to their body weight. Children stay in the programme until they reach their target weight (15% increase in weight from admission or, >85% WFH, absence of oedema), which usually takes around eight weeks.

For children suffering from complications associated with SAM and/or no appetite and for infants

Origins of the 'partnership' approach

Many NGOs responding to emergency situations from 2003 to 2005 using the CMAM approach had highly successful programmes. However, frustration arose when NGOs attempted to 'hand over' to the MoH once the acute emergency phase was over. Attempts to achieve full transition to MoH-led service provision appeared to fail. This was most likely due to the nature of emergency programming, where substantial inputs were required to rapidly scale-up response to a deteriorating situation. Emergency programming necessarily took on a vertical approach, as it required cars, staff, stores, supplies, etc, in order to reach the objectives of 'saving lives' in a timely fashion.

After the peak emergency period was over, while capacity would certainly have been built within the MoH through trainings, skill development, etc, the vertical nature of programming generally meant that the MoH were not the lead partner, or driver of the response. It was then often challenging for the MoH (in a resource-constrained environment) to assume responsibility for a programme which had not been 'theirs' in the first place.

The reaction to this frustration was to devise a new strategy of using a partnership approach. This involved establishing CMAM services (particularly OTP) in 'non-emergency' times with the focus firmly on 'ownership' by the MoH, i.e. looking at involving the MoH in longer-term programming from the start. It was envisaged that this would not only establish a much needed service during 'normal' times, but would also provide a capacity base from which services could be rapidly scaled-up at times of food crisis. Programming that viewed sustainability as the primary objective necessarily required that the thinking move away from traditional emergency dominated approaches, to one of partnership. In doing so, the nature of the dialogue between the partners was the most crucial element in defining the relationship dynamics that in turn determined programme success/failure. N-CMAM developed partnerships with:

  • The MoH. Formal 'Memoranda of Understanding' (MoU) were regularly signed between the MoH/Regional Health Bureau (RHB) and CONCERN in the four main regions. These MoUs proved to be very important documents, as they transparently outlined roles, responsibilities and expectations of each of the partners. In particular, they described CONCERN's role as that of technical assistance, rather than that of direct implementation. The MoUs also directed the expansion of services within the regions according to the needs and interests of the MoH.
  • UNICEF. UNICEF is mandated to support government health services for management of SAM. They have acted as centralised Ready to Use Therapeutic Food (RUTF) procurers and suppliers to the RHBs.
  • Valid International. As the originator of CMAM (originally Community Therapeutic Care (CTC)) approach and long standing technical partner of CONCERN. Co-founder of the N-CMAM programme.
  • Other NGOs. Through coordination fora, provision of training and support for CMAM-related activities, on request.

This partnership approach was in line with CONCERN's global strategy, which reflects a move away from direct implementation towards working through local partners (for NCMAM, the lead partner being the MoH).

Ownership, Commitment, Integration = Sustainability

Several words can be used to describe processes associated with sustainability: integration, ownership, commitment, all of which can be hard to define and are prone to subjectivity. It is important to note that these terms often mean different things to different people in different contexts. Children suffering from SAM is an emotive subject, undoubtedly due to the elevated endemic levels seen in countries such as Ethiopia, and its associated high mortality rate if left untreated (or poorly treated). As a result, views on the most appropriate strategies for treatment and acceptable standards of programme quality can be somewhat contentious.

The term 'ownership' became something of a mantra for the N-CMAM programme, with the understanding that "ownership equals internalisation, i.e. is self-maintaining".2 Experience has also identified that the term 'commitment' is often the single most important element that determines whether OTP services can be successfully established, and whether they will continue on or not. However, even if commitment is high at OTP service delivery level (health centre or health post), each part of the health system has to be functioning on the most basic level, in order for the service to be successful. For example, if the logistic system is poorly functioning resulting in RUTF stock-outs, the OTP service will collapse; 'no product, no programme' describes it well. So commitment/ ownership, while the key element, is not enough on its own. It is important to note that this is not unique to SAM programming. The attempted integration of a new health/nutrition initiatives into routine health delivery require that the system is functioning at least at moderate level, in order to succeed.

Table 1 summarises the timeline of key events in the evolution of acute malnutrition management in Ethiopia. Table 2 summarises the timeline in N-CMAM development.

Table 1: Key events timeline in Ethiopia
2000 First CTC/CMAM piloted in Ethiopia (Wolayita and Hadiya Zone, SNNP Region), Valid International with CONCERN and Oxfam.
2003 CTC/CMAM Research programme, Valid International/CONCERN, South Wollo Zone, Amhara Region - testing the efficacy and safety of the CTC/CMAM approach.
2003/4 Food crisis developed across many areas of the country. Many INGOs moved to programming using the CMAM approach, but coverage of services was low and only in certain areas. MoH-led evaluation of the emergency response identified an urgent need for rapid scale-up of SAM services.
2004 First national guidelines for the management of SAM introduced, focused on in-patient treatment.
2005 Demographic Health Survey (DHS) estimates national prevalence of acute malnutrition (children under-5) unchanged at 11%; stunting prevalence reduced by 5%, to a national rate of 47%.3
2005/6 HSDP-3 launched. Roll-out of Health Services Extension Package (HSEP) to be implemented through a cadre of 35,000 female community-based Health Extension Workers (HEWs).
2005 -2007 Relative period of food security across most of the country, although pockets of malnutrition remained. Screening figures from the new EOS programme (2004 onwards) highlighted very high levels of endemic acute malnutrition.4
March 2007 MoH endorsement of the revised national SAM guidelines, including extensive description of OTP and community mobilisation.
February 2008 Ethiopia co-hosts the launch of Lancet series on Maternal and Child Undernutrition. The National Nutrition Programme (NNP) for Ethiopia also launched5, both events providing evidence of the changing policy environment; with nutrition emerging from relative obscurity to the forefront of the policy agenda, at both national and international level.
April 2008 Attendance of State Minister of Health at international CMAM conference in Washington DC. This was an important event because Ethiopia's leading role in the continuing history of CMAM development was reiterated, particularly with regard to government-led programming.
2008/9 Food crisis developed across many areas of the country. OTP services rapidly decentralised down to health post (village) level across the four main regions of the country.6


Table 2: N-CMAM programme timeline
July 2005 Grant awarded from the US Office for Disaster Assistance (OFDA) to pilot an integrated partnership approach with the MoH for scale-up of SAM services. Grant to CONCERN, sub-grant to Valid International - project implemented together in light of synergistic relationship. Valid providing the ideas/innovation, CONCERN providing the historical base and well-established systems that allowed for programming in Ethiopia. The programme was implemented jointly for two years, after which CONCERN continued alone.
Oct 2005 Meeting with UNICEF with informal partnership formed. UNICEF to continue scaling-up support for the MoH to establish in-patient services (with some support for OTP). CONCERN/Valid to focus on supporting the MoH to establish OTP in selected operational areas.
2006/7 TFP working group formed (MoH, UNICEF, CONCERN, and Valid International). Development of 'TFP strategy paper', which outlined the strategies and guiding principles of national scale-up of SAM treatment services. It included a summary of key elements that must be in place in order to realise the scale-up of quality SAM services, such as training, reporting, supervision, community mobilisation, etc.
2005 - early 2007 Early stage of the programme, with considerable time spent in building-up relationships with key MoH figures in all four regions. OTP services commenced first in five health centres in Jimma Zone, Oromia region. Slowly increasing number of OTPs established at health centre level (from 0 to 37 HCs in 21 woredas). Through learning and experience, the 'minimal support' technical package was developed (see later), with time/energy also devoted to reviewing, analysing and developing Concern's internal CMAM strategy for Ethiopia.
2006 - 7 In collaboration with UNICEF, substantial support given to the MoH in developing the revised national SAM guidelines, which included OTP and community mobilisation activities.
May 2007 - January 2010 Second grant from OFDA enabled rapid scale-up of CMAM services. The N-CMAM programme began to gain influence, partly because of scale-up (from 37 to 187 health centres/posts across 87 woredas) as bigger programmes have louder voices, but also partly due to increasing engagement in what was going on incountry, e.g. input into the revision of national curricula for nutrition, providing the training manual that was adapted for the HEWs for OTP decentralisation, etc.
2008 - 9 Considerable scale-up of CONCERN's emergency nutrition activities (see Box 3). Support given from N-CMAM programme such as budget for supplies, trainings, vehicles, personnel, etc.


N-CMAM Programme Goal

The goal of the N-CMAM programme is to reduce morbidity and mortality associated with SAM in Ethiopia, through supporting the MoH to scale-up quality CMAM services (particularly OTP) within the routine health delivery system; that continues to function over the longer-term and can therefore be rapidly expanded during times of food insecurity.

Box 2: Outline of the National Nutrition Programme (NNP)

The objective of this programme is better harmonisation and coordination of various approaches to manage and prevent malnutrition. The NNP aims to reduce the levels of stunting, wasting, underweight and low-birth-weight infants, thus contributing to Ethiopia's efforts to reach the relevant Millennium Development Goal (MDG) indicators by 2015. The NNP consists of two main components:

  1. 'Supporting Service Delivery' which includes 'increased access for the management of SAM.'
  2. 'Institutional Strengthening and Capacity Building'.

The development of the NNP should be considered a major achievement for Ethiopia, as the country endeavours to tackle its long-standing problems with malnutrition. Rigorous evaluation of the NNP will be essential, to maximise the learning from this initiative.

N-CMAM activities and the 'minimal support' package

Sometimes a sibling is given responsiiblity for bringing a child to OTP

In addition to the support given to CONCERNs emergency interventions (see Box 3), the N-CMAM team assisted the MoH in selected areas of the four main regions (Oromia, Tigray, SNNP and Amhara) to establish and maintain OTP services. As UNICEF had been supporting the set-up and monitoring of in-patient units (TFUs or SCs) since 2003, the N-CMAM programme focused mostly on OTP service provision and community mobilisation. However, where gaps existed, the team also supported the set-up of in-patient services. From 2007 onwards, expansion of services was to areas identified by government authorities as most in need, usually determined by early warning data, identification as 'priority hot-spot area'7 and EOS screening data.

Box 3: Emergency programming

During the food crisis of 2008/9, CONCERN scaled-up to emergency nutrition response, implementing the 'full' CMAM package (that included supplementary feeding support, OTP and SC set-up) in 10 woredas. In addition, blanket supplementary feeding distributions were implemented in selected woredas of SNNP, Amhara, Tigray and Oromia regions, in attempts to contain rapidly deteriorating situations.

Total numbers of beneficiaries treated through the programmes during 2008/9:

Intervention Target group Total number of beneficiaries
Blanket supplementary feeding (SF) (1 month SF ration provided) Children 6-59 months 122,361
Pregnant (3rd trimester) and lactating women (infant <6 months) 20,989
Targeted SFP MAM children 6-59 months 54,943
MAM pregnant (3rd trimester) and lactating women (infant < 6 months) 30,069
OTP 11 11,689
TOTAL 19 240,051


The decision to scale-up to the emergency response was made either due to rising admissions to OTP (in areas that were being supported by the NCMAM programme), or when the results of rapid nutrition assessments or standard nutrition surveys indicated a worsening situation. Decisions for phase-out of emergency interventions were only made once results of standard nutrition surveys indicated that the vulnerable population had sufficiently 'recovered' (according to the Ethiopian Disaster Prevention and Preparedness Agency (DPPA) 'classification of malnutrition').

Programmatic results mostly reached Sphere recommendations, although some areas experienced poor recovery rates (see field article on Dessie Zuria in this issue, 'The history of nutrition in Dessi Zuria').

Along with 25 standard nutrition surveys and 10 Rapid Nutrition Assessments, a total of five CSAS coverage surveys to monitor the quality and impact of these selective feeding programmes were conducted. CONCERN placed a strong and continued emphasis on coverage assessments, due to the importance of measuring and quantifying the levels of service uptake by the target population. The reasons why some areas had high uptake and others low could be understood and lessons learned from both scenarios. Coverage results well exceeded the Sphere recommendations (>50% coverage in a rural area) in all but one assessment (where OTP point coverage was estimated at 46.4%, Dessie Zuria woreda October 2008).

A key challenge for CONCERN's emergency interventions was how to scale-up and then, crucially, scale-down again, without undermining the work that had been previously done through the N-CMAM capacity building approach. The period of transition back to N-CMAM's 'minimal support' approach could be interesting, as partners were required to realign expectations of assistance once the acute emergency period was over. Phasing-out workshops helped to reinforce the understanding of roles and responsibilities during the transition period, where OTP and SC services were again fully managed by the MoH. Learning reviews were also conducted at the end of each emergency intervention, in order to maximise learning for future interventions.

The 'minimal support' package was developed in consideration of the most important areas where the MoH needed support, to establish and then maintain quality services. The approach used a process of incremental capacity building to ensure that the ownership of the programme was always firmly in the hands of the MoH. The minimal support package included the following activities for set-up of services and provision of ongoing support:

  • OTP theoretical and in-patient SAM case management trainings, particularly at startup of services
  • Regular on-the job/refresher trainings
  • ToT trainings Joint supportive supervision and follow-up
  • Community mobilisation activities
  • Conducting pre- and post-workshop assessments
  • Facilitation of regular zonal and regional dissemination workshops/review meetings
  • Facilitation of experience exchange visits
  • In-patient training and support was provided on an ad hoc basis, when requested by the MoH, UNICEF or other INGOs.

It is important to note that the physical inputs in the minimal support package described above do not capture the relationship dynamics that must also be attended to, if ownership is to be achieved.

Community mobilisation and additional programme activities

Community mobilisation was a strong feature of the programme, with the focus on a more incremental version of community activation than the 'campaign-style' mass screening mobilisation common for emergency programmes. CONCERN's social development staff in consultation with the HEWs and other health staff utilised a strategy in 'new' CMAM areas as follows;

  1. Mapping: identifying key actors and identifying community structures already in existence, particularly those active in health/nutrition
  2. Discussing: commencing a public health dialogue with identified volunteers, including realistic time-frames of availablity to screen for SAM and conduct home-visits for follow-up.
  3. Training: 1 day training including sessions on underlying causes of malnutrition, use of MUAC and importance of defaulter followup.

As the focus of the N-CMAM programme was on innovation and learning, programme activities have included conducting a number of operational studies. These include a defaulter tracing study regular social development reviews and strategies, health system review, cost-analysis and effectiveness and investigation into the role of traditional practitioners in identification and referral of SAM. Case studies and examples of 'best-practice' were collected, along with programme learning reviews and evaluations.

At the request of the MoH, N-CMAM activities also included ToT in SAM case management, participation in multi-agency food security/ post-harvest needs assessments, support for EOS screening campaigns, support for additional monitoring & evaluation/ supervision activities, extensive attendance at national and regional coordination meetings and involvement in national initiatives, such as the recent Landscape Analysis conducted in Ethiopia.

The N-CMAM team worked alongside CONCERN's health and nutrition unit to conduct numerous assessments. From 2008 t o 2010, a total of 25 standard nutrition surveys and 10 rapid nutrition assessments were conducted, mainly to inform phase-in or phaseout of emergency nutrition activities. Additionally nine coverage surveys using centric systematic area sampling (CSAS) methodology were conducted.


The N-CMAM team relied on reporting channels of the MoH which uses a particular reporting format (developed by UNICEF) that in turn feeds into the centralised database for TFP reporting which is held at Addis Ababa level. However, as these reports are not yet integrated into the regular Health Management Information System (HMIS) reporting system of the MoH, TFP monthly reports were sometimes incomplete and were rarely sent in a timely fashion to regional level. It was often necessary, through a mixture of supervisory visits and reminder telephone calls for N-CMAM staff to collect 'missing' reports. The reporting rate is therefore estimated to be between 75-90% for the statistics presented here.

From January 2006 to December 2009, across the four main regions, 40,899 children with SAM were admitted to OTP services established at health centres, see Figure 1.8 A steady increase in rate of admissions was observed, with the expansion of OTP services within health centres across the four regions.9 Rising admissions could also be due to the higher levels of food insecurity seen in parts of the country where N-CMAM was operational during 2008/9. Table 3 and Figure 2 describe NCMAM programme performance, with a total of 31,480 discharges from January 2006 to December 2009.

Table 3: N-CMAM programme performance data Jan 2006 - Dec 2010, four regions
Year Jan-Dec 2006 Jan-Dec 2007 Jan-Dec 2008 Jan-Dec 2009
Total number of discharges (n) 1,343 4,463 10,171 15,496
Cured (%) 56 61 66.9 77.3
Death (%) 1 0.9 1.2 1.1
Unknown (%)10 0 11.6 15.1 9.9
Defaulter (%) 35.5 20.3 12.3 7.4
Non-responder (%) 0.5 3.2 3.1 3.2
Medical transfer (%) 7.0 3.0 1.4 1.1
  100% 100% 100% 100%


Shemsiya (2 years) failing the RUTF appetite test

Table 3 and Figure 2 describe N-CMAM programme performance, with a total of 31,480 discharges from Janurary 2006 to December 2009. 0Table 3 and Figure 2 demonstrate very encouraging programme results. The recovery rate steadily improved from 56% in 2006 to 77.3% in 2009, thus exceeding International Sphere recommendations for >75% recovery rate.11 Most of these gains in recovery are a result of declining defaulter/unknown rates and to some degree, the rate of medical transfer. The trend of improved recovery rates clearly demonstrates the MoH becoming more confident in their ability to offer quality services, with minimum technical support. It also suggests children are less likely to default, with improved access to services and increasing satisfaction with the service.

While an overall declining trend of defaulter and unknown rates was seen, they remained high. This is largely due to the challenging topography in many areas of the country and logistical problems of ensuring that all health facilities had an uninterrupted supply chain of RUTF. The mortality rate was consistently low. However as the number of unknown cases remains high, it is possible that some mortality occurred within this category. As expected, transfer rates to in-patient care reduced as access to services increased. Where children were identified earlier in their disease process, they generally had fewer medical complications and may not have reached the stage where they lost their appetite. This meant that burden on in-patient services was further reduced and minimised family and household disruption.


One of the major principles of CMAM is 'increased coverage' of services. During 2010, four coverage surveys using the CSAS methodology were conducted to assess coverage of the national TFP roll-out in the four main regions.12 Two of the surveys were conducted in areas that had received N-CMAM support. The results were impressive, particularly for period coverage which exceeded the Sphere target of > 50% in a rural area.13 Point coverage estimates were lower, highlighting the importance of continued efforts in community mobilisation (see above for a description of community mobilisation activities).

During 2009/10, N-CMAM, in collaboration with Tigray RHB, piloted a new methodology for assessing coverage of OTP services. This 'Semi Qualitative Evaluation of Access and Coverage' (SQUEAC) methodology was developed as a less resource-intensive (and therefore more 'MoH friendly') way of evaluating programme coverage and identifying barriers to service access and uptake.14 Between October 2008 and April 2010, three rounds of SQUEAC were implemented, with coverage steadily improving with each round (half of health facilities reaching >50% by the third round). Encouragingly, Tigray RHB not only took the lead by the third round, they also supplied the budget for the survey, with CONCERN providing technical assistance only.15

Training of MoH health professionals and community members

Table 4 outlines the focus that was placed on training of health staff. It is now well-known that OTP success is contingent on a responsive and well-informed community, considerable effort and dedicated resources were therefore also employed for Community Mobilisation (CM) training, using Volunteer Community Health Workers (VCHWs). To ensure trainings were conducted in a standardised manner, NCMAM staff developed a training manual for the management of SAM during 2006-7, according to National Guidelines. The version has been improved throughout the programme's life cycle, with CONCERN taking a major role in the development of OTP training materials used at national level.

Table 4: Training data January 2006 - December 2009, four regions
Year Health professionals Health Extension Workers HEW Supervisors16 VCHW
2006 108 42 0 565
2007 380 359 0 1.339
2008 547 1,236 113 1,965
2009 534 1,593 284 2,161
TOTAL 1569 3230 397 6030



By 2006, CMAM/CTC was still relatively new to Ethiopia and had limited coverage, so a twoday national workshop was hosted in Addis Ababa to bring together the MoH, UN agencies, donors and NGOs. The objective was to discuss the current status of therapeutic care in Ethiopia and formulate plans for increasing coverage of quality services. From 2007, the NCMAM team recognised that the facilitation of workshops/review meetings could be a valuable tool for improving the quality of services. MoH staff involved in OTP implementation were brought together in the same room as key decision makers from regional, zonal and federal level MoH, to discuss the challenges faced and devise appropriate solutions (Table 5). These workshops (while expensive and time-consuming to prepare for and conduct) provided excellent opportunities both for developing coordination and encouraging accountability among partners. Additionally, 'phasing-out workshops' (after emergency nutrition interventions, see Box 3) proved important fora to reinforce the understanding of roles and responsibilities, including the re-design of action plans for future activities when OTP was transferred back from more intensive 'emergency' level support provided by CONCERN to full MoH management.

Table 5: Summary of N-CMAM regional review meetings/workshops by year
Year No. of workshops No. of regions covered Total number of participants
2007 5 4 (SNNP, Oromia, Amhara, Tigray) 262
200817 3 2 (SNNP, Oromia) 92
2009 11 4 (SNNP, Oromia, Amhara, Tigray) 897
TOTAL 19   1251


Experience-sharing visits

Shemsiya with her family

A learning and information-exchange visit by Ethiopian MoH officials to Malawi was facilitated during 2006, with representatives from UNICEF joining the trip. The goal was to learn and share lessons on the treatment of SAM as part of routine health services. The visit provided an excellent opportunity for Ethiopia staff to learn from Malawi's recent national scale-up experience, hearing first-hand about the successes and challenges that the Malawian MoH had faced.

From 2008 onwards, a strategy of experience- sharing visits for MoH staff to other regions within country was employed. Six visits were organised with a total of 181 MoH staff travelling. Health workers who were performing well were invited to travel to areas in need of encouragement, to improve programme performance. The success of these visits hinged on the fact that advice was being given by the MoH to the MoH - not from CONCERN - which gave added weight to recommendations of how to improve service delivery. The visits appeared to act as powerful motivational tools, with potential for longerterm impact.


While the programme was always results orientated, it was viewed through the lens of public health. This focus enabled it to keep going through the early years of low recovery rates (only 56% in 2006), in the belief that if services were incrementally established at-scale, improvements in service quality could follow. This was as long as the MoH was provided with sufficient, appropriate and well-targeted support.

At present, the only benchmarks for SAM treatment programmes are provided through the Sphere Project. While reaching these recommendations should always be the aim of the programme, it must be remembered that they have evolved from humanitarian emergency programmes that are generally implemented by well-resourced INGOs. It could be useful if suitable benchmarks for MoH-led national scale-up programmes that provide appropriate reference points for programme quality, especially during the early years of implementation were developed.

A number of key factors/processes, listed below, promoted an enabling environment within which the N-CMAM programme could achieve its objectives.

a) Continuous and sufficient funding from July 2005 to date, through a series of grants from OFDA and CONCERN-sourced funding, which:

  • Allowed for programme scale-up and continuation, with the ability to deliver on commitments.
  • Enabled regular processes of learning, monitoring and evaluation, with budget allocated for operational studies and innovative programming approaches.
  • Offered dedicated budget for senior staff and an expanded team at capital level that enabled rapid response to requests for technical assistance from partners (MoH, UNICEF, other NGOs), attendance at a wealth of coordination fora and for staff to take an active role in policy developments.
  • Assisted the scale-up to emergency programming, where required (see Box 3).

b) Strong focus on training of the N-CMAM team in how to support partner staff, rather than undertaking direct implementation. Building the capacity of CONCERN's partner required the team to deliver support from the sidelines, often a more difficult task than direct implementation.

c) Strong and continued focus on community mobilisation, with dedicated Social Development staff. This helped to create the 'demand driven' version of OTP, whereby the rapid and visible recovery of children acts as a profound motivational force for parents, health workers and the wider community.18

d) Initial targeting of districts that had experienced limited INGO support, in an attempt to avoid areas where dependence on NGOs for nutrition-related programming was highest. Additionally, CONCERN invoked a principle of not having a full-time presence in operational areas. Instead, CONCERN had a base in the capital that could provide support and dialogue with the woreda and Regional MoH, but no permanent presence.

e) The MoUs that were regularly signed with RHBs supplied the all-important official authentication for programmatic aims and objectives.


During programme implementation, a number of challenges were experienced. Many of them were common to overburdened health systems, but some were particularly accentuated with CMAM.

Programme specific challenges

Welcome message by Ato Yohannes, the Deputy Head of Tigray RHB.

Transport continues to be a major challenge for the MoH across this vast country. Where vehicles exist, they are often in need of maintenance and lack budget for fuel. OTP is a logisticallyheavy service modality due to the bulky product (RUTF). Ensuring sufficient and adequate storage space also presents major difficulties. Additionally, the service is not a 'one off' - caregivers need to come back each week for eight weeks on average, requiring the OTP to have regular, uninterrupted supplies to ensure successful outcomes.

Facilitation of sufficient and adequate supervision at-scale was problematic, especially considering the transport challenges described above. While the supervisory visits were always well appreciated by health staff, due to many competing commitments, it was not easy for over-burdened MoH supervisory staff to make enough time for visits to health facilities. A standard supervision checklist for OTP was developed during the programme (in collaboration with UNICEF and the MoH), and is currently being integrated within the routine health supervision system.

Collection and collation of programme statistics was a major challenge. The data presented above is only from health centres. Since the national 'roll-out' of TFP (2008 onwards) and the decentralisation of OTP services to health post level, staff providing regular and accurate reporting face many additional hurdles.

High staff turnover within the MoH was the primary obstacle to ongoing knowledge retention at health facilities, resulting in untrained staff often found implementing OTP. Focal persons for nutrition to co-ordinate and monitor activities at regional, zonal and district level were appointed. However, the high rate of staff turnover meant that it was important to involve all members of the health team from the start, to minimise the risk of the programme collapsing if one person left.

Routine antibiotic treatment was mostly not available at the health facilities as they are supplied through a delivery system that is different RUTF supply.

Community mobilisation mechanisms were not always clear and readily available for use, resulting in a parallel system needing to be setup (although the programme collaborated with whatever mechanisms did exist in the community). The UNICEF-supported Community Based Nutrition (CBN) programme now presents an excellent opportunity to integrate SAM screening into the growth monitoring activities carried out by VCHPs, but the modalities of this have yet to be agreed.

Challenges in the environment

Integration of scaled-up OTP as part of routine health delivery provides many challenges for an overburdened and under-resourced health system, including the time taken to actually treat the children, attendance at trainings, reporting, ordering, transport and storage of supplies, etc.

Competing priorities within the health system are also a challenge, since many health/HIV related initiatives are being 'rolled out' at the same time in Ethiopia. While improvements in health services are obviously welcome, it does exert pressure, e.g. attendance at trainings can result in health staffs spending considerable time away from facilities. Additionally 'per-diem' rates risk becoming competitive, with initiatives that pay higher rates appearing more attractive.

Lessons learned

The length of time that the process of integration requires on many levels. Continual dialogue and interaction with key actors is needed to ensure objectives are being met and opportunities capitalised on.

Development of acceptance at community level. While RUTF offers 'instant' solutions (sometimes described as the 'steroid treatment' for the dramatic results), understanding by the community of the causes of malnutrition and appropriate health seeking behaviours to treat and prevent it, takes a much longer time to mature.

The importance of using both formal and informal channels of communication and networking to muster influence. As a relatively small NGO, we (unsurprisingly) had limited access to key decision makers. It was important therefore to utilise all available openings in order to advocate for OTP uptake and decentralisation. Informal channels proved remarkably effective in gaining access to important actors of the policy making process.

Personality driven gains. The effect of having key people in key places in the MoH, e.g. a motivated zonal health head, should not be underestimated. Having certain individuals in key advocacy roles is also important, for example, the previous country representative of UNICEF was a nutritionist. While the reliance on key people (or one key person in an area) was undoubtedly a high risk strategy, there was often little alternative. Additionally, while the person/s was in place, opportunities could present themselves, providing unexpected advantages.

Sufficient and adequate supervision is difficult to achieve on a large scale. Joint supervision is essential and must be insisted upon to achieve maximum effect. Improvement in service provision is only likely if the health workers are convinced that more senior staff are interested enough in programme results. Initially it might be necessary to conduct additional supervision to ensure service quality (especially when the service is new). However, integration into routine health service supervision must be aimed for if sustainability is to be achieved.

N-CMAM Programme limitations

Although a number of 'strategy meetings' were held, a strategy as such was not developed. This was partly due to the short-term funding cycles of the programme that presented some barriers for the articulation of a strategic vision. Longerterm funding for N-CMAM proved elusive, despite attempts to secure it.

While the N-CMAM is a much more costeffective version of OTP than 'usual' NGO supported versions, the programme remains relatively expensive. Considerable budget is required for trainings, cars for supervision, staff salaries, etc.

N-CMAM prioritised attention on OTP over the establishment of in-patient services. This decision was taken for two reasons. First, existing UNICEF support for in-patient services (described above) and second, with limited CONCERN capacity, the focus (as public health advocates) should be on the 95% of children who could be treated in OTP. It is acknowledged that alternate views exist, that might consider N-CMAM should have prioritised inpatient care also.

There were a number of missed opportunities. For example, linkages with preventative nutrition interventions were very limited, as were linkages with hygiene and other sanitation activities. Health education at the OTP was irregular and could be sub-optimal.

While the programme's aims and objectives have always been to support the MoH, it is acknowledged that this particular version of NCMAM remains NGO-driven, where sustainability after phase-out is not guaranteed.


The N-CMAM programme started from small beginnings and grew into a relatively large and successful intervention. It capitalised on the national and international momentum that was building around nutrition issues. It contributed to the national roll-out of SAM treatment services that is now underway in Ethiopia, by demonstrating that the MoH were well able to manage and implement OTP services of increasing quality at-scale. It reflected that the process might take some time and considerable expense.

The national roll-out of the TFP has been subject to considerable effort from the MoH (supported by UNICEF and other partners). To date, approximately 30% of Ethiopia's health facilities are offering OTP services, an impressive achievement over a short period of time. The challenge now will be how to maintain quality of the OTP, when implemented at scale. The difficulties of adequate monitoring, supervision and reporting are enhanced, as is the logistic burden for the health service. Strategic planning, along with strong leadership from the MoH will be required, in order that the TFP rollout can be strengthened and maintained.

For further information, contact: concern.ethiopia@concern.net

1WHO, UNICEF and SCN Informal Consultation on Community-Based Management of SAM in Children, Geneva, 21-23 November 2005

2Handy, C. Understanding Organisations (4th Ed) 1999. Penguin.

3Ethiopian DHS 2005

4While the quality of EOS screening figures was not assured, it was the first time that screening for malnutrition had been conducted at-scale; the very high absolute numbers of children <5 years with acute malnutrition provided policy-makers with a powerful advocacy tool.

5The NNP has been designed to translate the National Nutrition Strategy into a 10- year action plan in two phases; the first phase is for 5-years 2008-2013, at an estimated cost of $370 million.

6Chamois, S. 'Decentralisation of out-patient management of severe malnutrition in Ethiopia', Field Exchange, ENN, July 2009, Issue 36.

7The Emergency Nutrition Coordination Unit (ENCU) produces a 'hot-spot matrix' identifying priority areas in need of nutritional support, which is revised every six months.

8The programme continues on into 2010 but full results will not be available until early 2011.

9Figures presented here are for health centres as 'roll out' to health post occurred gradually during the year 2009 in NCMAM areas. From 2010 onwards, programme figures will be collected from health posts.

10During the 2007 revision of the national SAM guidelines, defaulters were divided into 'unknown' i.e. unconfirmed defaulters and 'defaulter' i.e. those defaulters confirmed through a home visit. The values need to be added together for the total default rate.

11The Sphere project recommends: recovery > 75%, death <10%, default < 15%, coverage > 50% (in rural areas)

12The CSAS assessments were coordinated by the Ethiopian Health and Nutrition Research Institute (EHNRI) and the MoH. CONCERN provided the technical support for implementation of the surveys, which were funded by UNICEF.

13Mareko woreda, Gurage zone, SNNP Region: OTP period coverage 60.9% (95% CI: 52.8%-68.6%) and point coverage 37.8% (95% CI: 28.2%-48.1%). Taytaw Machew woreda, Tigray Region: OTP period coverage 56.2% (95% CI: 44.1% - 67.8%) and OTP point coverage 27.3% (95% CI: 15.0% - 42.8%).

14Myatt, M. 'SQUEAC: Low resource method to evaluate access and coverage of programmes', Field Exchange, ENN, June 2008, Issue 33.

15Schofield, L, et al. 'SQUEAC in routine monitoring of CMAM programme coverage in Ethiopia'. Field Exchange, ENN, March 2010, Issue 38.

16Recognising that the new cadre of HEWs were in need of supervision, the MoH trained and deployed HEW supervisors from 2008 onwards.

17Fewer workshops were conducted during 2008 as many of the N-CMAM staffs were involved in Concern's scale-up of emergency programming (see Box 3)

18'Temporal Integration' Demand driven CTC. Steve Collins, Emergency Nutrition Network (ENN) Special Supplement Series, No. 2, November 2004.

Imported from FEX website


About This Article

Article type: 
Original articles

Download & Citation

Recommended Citation
Citation Tools