Quantity through quality: Scaling up CMAM by improving programmes Access

By Saul Guerrero & Maureen Gallagher

Saul Guerrero is the Senior Evaluations, Learning and Accountability (ELA) Advisor at ACF UK based in London. Prior to joining ACF, he worked for Valid International Ltd. in the research, development and roll-out of CTC/CMAM. He has worked in over 18 countries in Africa and Asia.

Maureen Gallagher is the Senior Nutrition & Health Advisor ACF USA based in New York. She has worked for the last 10 years in nutrition, food security and hygiene promotion programming in Niger, East Timor, Uganda, Chad, DRC, Burma, Sudan and Nigeria.

The authors would like to thank nutrition advisors across ACF International for their direct contribution, in particular Anne- Dominique Israel, Sandra Mutuma, Cecile Basquin, and Silke Pietzsch. We would also like to thank all ACF nutrition advisors on their contributions in discussions on the issue including Elisa Dominguez, Marisa Sanchez, Nyauma Nyasani, Marie-Sophie Whitney, Olivia Freire, Fabienne Rousseau and Maria Masferrer. Our gratitude also goes to all the ACF International field coordinators and MoH staff for their on-going involvement in reviewing the past, present and future of CMAM programmes.

Background

ACF International’s strategy for scaling up nutrition programming is underpinned by one simple idea: we are not reaching enough of the affected population. According to the 2008 Lancet series, only between 5%¹-10%² of children suffering from acute malnutrition are receiving nutritional care. To deal with this, ACF International has committed to a gradual process of scaling up interventions to reach an estimated 500,000 children per annum by the year 2015. In 2010, the organisation reached just under 225,000 children through its programmes. To meet its objectives, the organisation must effectively double its current caseload. To achieve this, ACF has placed considerable emphasis on securing the necessary political will and civil society participation at national and international level to enable growth and expansion of services³. Internally, it has also placed great emphasis on partnerships and capacity building as a means of enabling programmes to expand and reach new geographical areas.

All in all, this represents an outward vision of growth, an approach that favours expansion over consolidation; the replication of existing approaches on the assumption that more of the same will deliver results. The implicit confidence in the performance of existing programmes is undoubtedly linked to the great strides made over the last decade with the decentralisation of care. By shifting from Therapeutic Feeding Centres (TFC) to CMAM, the organisation has laid the foundation for a significant increase in programme uptake. Yet, like many nutrition organisations around the world, ACF is gradually coming to the realisation that the shift in treatment approaches is no guarantee for success. There is now an increasing body of evidence showing that offering services, even closer to the communities, is not tantamount to improving access. Whilst the efficacy of the CMAM model and protocol is now firmly established, its effectiveness is still dependent on the quality of programme implementation by (or with the support of) NGOs, as often Ministries of Health (MoH) have limited resources for nutritional activities.

Scaling up must therefore start by consolidating our work, by finding ways of reaching those that we are consistently excluding. The potential benefits of consolidating our work, or shifting to an inward approach for scaling up, has been conclusively established. If we were to increase our current coverage by 30% we would not only meet international minimum standards, but we could reach our target of 500,000 children per year without opening a single additional programme. We do not yet have a generic recipe for effective CMAM programming, what we do have is sufficient evidence to start developing a new approach that focuses more closely on ensuring programme effectiveness. The aim of this paper is to use some of the available evidence to propose a clear and specific definition of effective CMAM programming. But it also seeks to go further. By reviewing results of current trends in programming, it sets out to identify key steps for consolidating ACF’s existing programmes, and to outline the programmatic and organisational transformations that these would have for ACF.

What’s in a number? Defining quality in nutrition programmes

The aim of public health nutrition interventions, such as CMAM, is to meet the needs of the largest possible proportion of an affected population⁴. There are two sides to the question of needs met. On the one hand, there is the quality of care or efficacy of treatment. This is generally assessed through standard nutrition indicators including cure, death, defaulter and non-responder rates. Although these indicators vary from context to context and according to a number of factors (including severity at presentation, level of compliance, implementation approach, etc), there is now ample evidence to support the idea that the efficacy of CMAM treatment protocols is close to 100%⁵ in a controlled environment. These efficacy indicators, however, give us only a partial view of the impact of the programme on the children reached. The other part of the needs met equation, the one often missing in our analysis, concerns the number of affected cases that programmes do not reach. The quality of nutrition programmes must be determined by a combination of the treatment efficacy outcomes vis-a-vis the proportion of the affected population being reached (coverage).

Coverage has been an integral component of humanitarian evaluative frameworks, with both OECD-DAC and ALNAP⁷ recommending that humanitarian agencies "…present an estimate of the proportion of those in need covered, expressed as a percentage, rather than an absolute number". The importance consistently attributed to coverage is exemplified in Figure 1. When programmes with coverage of 30% (A) successfully cure a high rate of children (B), the proportion of needs met is still low (C). When programmes with high coverage (X) cure only half of the admitted cases (Y) the proportion of needs met is higher (Z).

Since they first appeared in 2004⁸, the SPHERE Standards have included specific coverage indicators for nutrition interventions in rural (50%), urban (70%) and camp environments (90%). The timing of the introduction of these new standards was important. Prior to the introduction of the CTC model in the late 1990s, inpatient programmes rarely reached over 25% of the affected population⁹. But by 2001, as CTC (and its successor, CMAM) became more common, NGOs consistently showed that it was possible for programmes to reach up to 70% coverage. This was taken as a sign of the intrinsic quality of the approach, rather than as a partial reflection of the active and direct involvement of NGOs in its implementation.

Great expectations: ACF programme coverage performance against international standards

ACF has a strong history in supporting coverage estimations (see Box 1). Between February 2010 and February 2012, ACF has carried out 15 coverage surveys in 11 different countries, with more planned for the short and mid-term future. Whilst the contexts have varied significantly, regular coverage surveillance has provided us with two valuable areas of information: coverage diagnosis and programme diagnosis. Together, these two areas of information provide clear ideas of where the problems and the solutions lie, offering practical recommendations for improving/ensuring the quality of programmes.

 

Based on the coverage monitoring data collected since 2007 (Table 2), on average ACF programmes achieve coverage of around 30% of the total affected (SAM) population¹⁰; for every three acutely malnourished children in our areas of operation, only one receives treatment through our programmes. None of the programmes surveyed so far has met the SPHERE minimum standards (>50% for rural areas, >70% for urban areas, >90% for camps). The comparison against SPHERE standards may only be relevant to some of these programmes which were implemented (directly by ACF), under emergency conditions. The remainder of these programmes have been implemented in partnership with (i.e. indirectly through national MOH. The SPHERE standards represent a set of benchmarks that have proven difficult to attain by integrated programmes led by MoH. This has resulted in the commonly held belief that integrated CMAM programmes run by MoH are intrinsically limited by infrastructure and resources, and are thus naturally unable to meet previous CMAM outcomes and standards. Whilst integrated CMAM programmes are different, focusing on the need for new standards for integrated programmes (or provision of external inputs to support achieving standards) represents an easy way out. What this argument effectively does is remove the pressure from NGOs to understand the factors affecting the performance of integrated CMAM prog- rammes. This in turn often leads to the creation of artificial programme conditions - including additional staff, financial incentives and supply systems - that help boost performance, but does little to strengthen local health systems.

Table 2: Coverage of ACF-supported nutrition programmes (2010-2012)
Country	Location	Date	Coverage
Burkina Faso	Tapoa	Feb 2010	21.8%
Chad	Kanem	Dec 2010	27.1%
Burkina Faso	Tapoa	Mar - Apr 2011	17.6%
Mauritania	Guuidimaka	Mar - Apr 2011	33.0%
Liberia	Monrovia	Feb - Apr 2011	24.8%
Mali	Gao	Jul 2011	35.4%
Nigeria	Yobe	Aug 2011	33%
Chad	Kanem	Sep 2011	36.4%
Chad	Bahr el Ghazal	Oct 2011	34.1%
South Sudan	Gogrial West	Oct 2011	44.7%
South Sudan	Aweil East	Nov 2011	45.5%
South Sudan	Twic	Dec 2011	27.3%
Myanmar	Maungdauw	Nov - Dec 2011	40.7%
Haiti	Haut Artibonite	January 2012	12.4%
Sierra Leone	Moyamba	February 2012	12.1%

 

The challenges faced by integrated CMAM programmes are very real: beneficiary populations, for instance, have a pre-conceived idea of what health facilities can and cannot offer, about the kind of (staff-patient) treatment they are likely to receive there, including official and/or unofficial costs of treatment. The location (often limited and sparse) of health facilities in a given area means that reaching those in need is generally determined by the proximity offered by the MoH infrastructure chosen as service delivery units for CMAM services. In addition, the resources (human and financial) available to MoH to carry out the supporting functions needed by any successful nutrition programme are very often not there. ACF has sought to address this, but the coverage performance of these programmes suggests that the current allocation of technical support and resources to support integrated programmes is based less on needs and gaps of health systems and more on the traditional organisational clinical expertise and focus on treatment efficacy. Meeting a higher proportion of the needs, and reaching the expected 500,000 SAM children a year, will require that ACF looks beyond clinical outcomes and addresses key factors for achieving high quality programmes.

Key factors for achieving high quality nutrition programmes

Part of the answer to the question of where support is needed to improve the quality and performance of our programmes is provided by the coverage assessments themselves. In 2007, the ACF Uganda nutrition team set out to prove that programme coverage could be increased through the strengthening of community mobilisation activities, including sensitisation, case-finding and follow-up. The programme succeeded in increasing coverage by more than 12% in 12 months¹². More recently, in December 2010, a coverage assessment carried out in Kanem (Chad) once again highlighted the need for improved community engagement. By taking on board the recommendations from this assessment, the programme was able to increase its coverage by over 9% in 10 months. Together, these two experiences prove that within a relatively short period, programmes can positively influence their coverage by addressing some of the bottlenecks¹³ affecting access and that community mobilisation can play a pivotal role in achieving this.

Since then, our understanding of the factors affecting programme performance has increased through available data from within and without the organisation. A review of 12 CMAM programmes published in 2010 concluded that programme coverage was directly affected by 1) the degree of rejection amongst referred children, 2) the level of awareness (about the condition and services) amongst the population, and 3) the distance between targeted communities and service delivery points¹⁴. The recognition and incorporation of mid upper arm circumference (MUAC) as part of national nutrition protocols for admission in many countries has significantly reduced rejections in programmes¹⁵. The other factors, however, continue to negatively impact programme performance. A pilot study carried out in 23 health centres by Concern Worldwide in Ethiopia identified lack of awareness about the programme as the single most important barrier affecting the performance of the integrated CMAM programme, preventing any of the facilities evaluated from reaching more than 50% of the affected population¹⁶.

ACF’s recent surge in coverage surveillance has created a body of evidence that corroborates these conclusions (see Table 3). In all coverage surveys carried out by ACF since February 2010, awareness about programme and/or malnutrition has been identified as the primary reason(s) for non-attendance. Simply put, the large majority of people in the communities where ACF works remain unaware of the existence of CMAM services, or do not perceive it as the solution to the condition affecting their children. The evidence suggests that the current approach to implementing and supporting integrated-CMAM programmes is inappropriate to deliver the promise of greater access that the CMAM model was built on. As the CMAM approach continues to be scaled-up and rolled out, it is more pressing than ever to revisit and review the current model involving health systems, communities and nutrition organisations. This does not mean starting from scratch; the wider public health sector has been tackling these issues for years, gathering valuable lessons and experiences that can be brought into the fold.

Rethinking the CMAM service delivery model and the role of nutrition organisations

The role of humanitarian agencies has shifted significantly since the direct interventions of the 1970s and 1980s. Organisations involved in public health programmes have gradually scaled up by working in partnership with local authorities. For nutrition organisations like ACF, this has represented a shift from direct implementation to "focusing on strengthening health systems’ own capacities to treat severe acute malnutrition"¹⁷. The health system strengthening approach - with nutrition as an entry point - often varies between nutrition organisations as well as between ACF missions¹⁸. Generally speaking, however, there are some fundamental areas for support:

• Supporting the coordination & creation (or review) of technical frameworks including national nutrition policy, protocols, guidelines, and training manuals, as members of National Technical Working Groups.
• Supporting strengthening capacity efforts (i.e. training, coaching, in some cases additional human resources (HR)) for staff involved in the management and implementation of CMAM activities, including national/regional/local managers, health facility staff and outreach workers.
• Strengthening/supporting supply chain management, including systems for forecasting, requesting and distributing essential drugs and/or ready to use therapeutic food (RUTF). Though these areas remain key to supporting the integration of CMAM into national health systems, the performance of integrated-CMAM programmes continues to falter. Addressing this requires a rethink of the present and future of the CMAM service delivery as part of the health system and the role of nutrition organisations like ACF in this process.

Consolidating experiences in health system strengthening

What is ACF’s CMAM health systems’ strengthening approach? With so many varying integration definitions and approaches between ACF missions, ACF is in the process of defining a clear position and model on the key factors that ensure ownership, performance and sustainability of integration of CMAM into routine health services. As health systems vary from country to country, one model cannot fit all, and it becomes important to learn from our existing experiences to see where we are, and to define a CMAM institutionalisation framework that is adaptable and replicable within different health systems. The emphasis of such a framework should consider all key elements for successful treatment yet maintain the existing health system as central to decision making.

Since 2009, a CMAM Integration Guide has been under development by ACF. The success of such a guide rests on its ability to capture experiences from across a wide range of contexts, and in particular, on its capacity to identify successful approa- ches and provide practical guidance in some key areas, including:

1. Operational Planning (e.g. what is the case load? How is it managed? How many days of treatment are provided and why? What incentive systems are in place and why?)
2. Human Resources (e.g. who is managing treatment? How are health workers involved in treatment? Are additional staff supported? How? Why?)
3. Logistics (e.g. who does the supply management? What is the medicine provision system?)
4. Training (e.g. what is the training approach? Who conducts training? How is training impact measured?)
5. Monitoring & Evaluation (M&E) (e.g. who does data collection and how? What is the health system personnel involvement and understanding of M&E?)

In collecting such information in a systematic manner, ACF can consolidate experiences, identify areas for further research/analysis, and develop an operational framework/key principles for integration including timeframes and exit strategies. In this process, factors will be addressed through the health system strengthening lens, in exploring how different components fit into the six health system strengthening blocks - service delivery, supply, health workforce, financing, health information systems, leadership & governance¹⁹. This framework will provide a clear position and strengthen the capacity of ACF in the shift from direct implementer to advisor on health systems’ strengthening. This requires looking at nutrition as a specific treatment in a larger public health setting.

Revising and prioritising community mobilisation

Raising awareness, sensitisation and social marketing have long been recognised as key components of successful public health interventions. The challenge for programmes operating through health structures is the limited or complete absence of financial resources allocated by most MoH to sensitisation and/or outreach activities. As a result, nutrition organisations like ACF have sought to replicate the same strategies used in NGOimplemented CMAM programmes, but without the financial compensation given to outreach workers. Instead, integrated CMAM programmes increasingly rely on existing community volunteers (generally linked to MoH) to deliver community mobilisation activities. Whilst this avoids creating parallel (and unsustainable) structures, this approach consistently faces operational challenges that ultimately define the (poor) performance of integrated CMAM programmes.

Improving community mobilisation to foster optimal programme coverage is less about addressing the individual challenges associated with working with volunteers²⁰, and more about redefining the overall paradigm that places individual community members (volunteers) and sustainability at the heart of a community mobilisation strategy. The current working model behind community mobilisation in integrated CMAM programmes is based on two fundamental assumptions: community volunteers are the primary means by which to identify and refer cases, and the activeness of these volunteers must be maintained without incurring payments which cannot be sustained by the local health systems. The biggest weakness of this model is that it overestimates the importance of individual volunteers and underestimates the importance of collective community involvement. The proportion of admissions in CMAM programmes between those referred by individual volunteers and self-referred by communities themselves is such that a different approach is possible.

Integrated CMAM programmes should aim to create a critical mass capable of triggering a more sustainable dynamic between the community and the services provided (see Figure 2). At the start of CMAM activities, efforts should be placed on large-scale community sensitisation and case-finding. This would lead to volunteers referring the majority of cases at the start (A). After a short period, motivation would naturally decrease leading to a drop in referrals by volunteers (B). In the meantime, the critical mass or momentum created by the rapid and visually clear recovery of SAM children would lead to a gradual increase in self-referrals (C). Over time, the number of cases that seek CMAM services spontaneously would overtake those referred by volunteers and can (assuming no significant barriers to access) lead to a sustainable and comprehensive model for ensuring programme coverage.

Achieving this dynamic would require, first and foremost, a prioritisation of community mobilisation activities as a key feature of the support provided by organisations like ACF. In practical terms, this would have implications for the profile of staff, and resources made available, to integrated CMAM programmes. Nutrition organisations like ACF should support MoH in the design, planning and implementation of sensitisation activities including mass media, traditional communication channels and the use of new technologies. It would also mean the involvement of nutrition organisations in the training and coordination of community outreach activities by volunteers. All efforts to increase community uptake of CMAM services, however, will need to be accompanied by the introduction of a service delivery structure capable of absorbing the increase in caseload, and capable of providing effective and appropriate care (including low waiting times, regular supply of RUTF, positive staff-beneficiary interface, high cure rates, etc). Experience has shown that it is this third element - capacity to deliver - that often proves problematic, in particular for integrated- CMAM programmes dependent on the availability and quality of existing health human resources and infrastructure.

Exploring alternative models of CMAM service delivery

The current model for integrated-CMAM programmes relies on the utilisation of health facilities for the delivery of treatment services. Based on this model, support organisations like ACF are tasked with identifying facilities capable of mainstreaming CMAM activities as part of their daily and/or weekly activities. The aim is then to introduce health system strengthening initiatives (e.g. staff training) designed to prepare these facilities for the arrival of newly identified SAM cases from the community.

This model represents a limited vision of health systems. Health systems can also include additional tiers such as Community Health Workers (CHWs). For many years, public health interventions tackling TB, HIV/AIDS, Malaria and Family Planning have turned to this tier for the delivery of support and care. Many of the reasons that have led to the decentralisation of care reflect the same challenges currently faced by integrated-CMAM programmes: weak health facilities with limited and overworked staff, high caseloads leading to long-waiting times, stigma associated with the condition and high opportunity-costs linked to attendance.

There is encouraging evidence that community case management (CCM) of acute malnutrition is not only possible, but can effectively deliver high quality results. Existing evidence from Malawi has shown that the outcomes of treatment delivered by CHWs are comparable to treatment delivered at health facility level²¹. Available studies have concluded that "home-based therapy with RUTF administered by village health aides is an effective approach to treating malnutrition during food crises in areas lacking health services".²² Similar large-scale research also carried out in Malawi²³ concluded that "home-based therapy with RUTF yields acceptable results without requiring medically trained personnel".

In 2011, Save the Children with the support of Tufts University, GAIN and Pepsico, carried out operational research on CCM of SAM in Southern Bangladesh. Unlike previous research, the Save the Children project measured both the efficacy of treatment and the coverage of the intervention. As in Malawi, the project achieved high recovery rates (92%), and low defaulting and mortality rates (7.5% and 0.1% respectively). The coverage of the programme (89%)²⁴ was found to be one of the highest ever recorded by a CMAM programme. Subsequent research has also shown that the project delivered high quality of care.

The Malawi and Bangladesh experiences show that a CHW-based service delivery model can be effective, but the evidence so far has been largely based in contexts with robust CHW networks or where additional resources have been made available to support these. The existence of a professional cadre of paid CHWs is no guarantee in itself. In 2005, the Ethiopian Government introduced the Health Extension Programme (HEP) designed to bring together all basic maternal and child health interventions, including nutrition. Yet, CMAM programmes have often struggled to incorporate nutrition activities (even case-finding alone) into a workload that initially included 17 different health packages, from HIV/AIDS to control of insects and rodents.

The evidence from Malawi and Bangladesh is promising, and the improved access and proximity offered by this approach could offer a way of decreasing pressure on health facilities (in high prevalence areas in particular), decreasing defaulting and improving programme coverage. More evidence is needed, from larger interventions, over longer periods of time and outside of the controlled environments (where there is large scale resource investment) of the Bangladesh and Malawi experiences. However, the real success of this type of CCM model may ultimately rest on the ability to implement such programmes as part of the much broader process of strengthening health systems and successfully linking CHWs, health facilities and the communities which they serve.

Conclusion

The shift away from centralised, inpatient care towards a community-based model was arguably one of the most important paradigm shifts in the history of public health nutrition. This shift, however, is far from complete; as nutrition interventions enter a new phase characterised by the integration of nutrition services into national health systems, the coverage and impact of these interventions is decreasing. Turning this around is possible, but to do so, nutrition organisations must adapt to the changing demands linked to health system strengthening and the prioritisation of community mobilisation/awareness. The question for organisations like ACF is not how to provide the same support in a different context, but rather, what kind of support does the new context require and how can this be provided. The answers to these questions are likely to fundamentally change nutrition support organisations - from their staff profiles to their strategic objectives - but in doing so it will make organisations better prepared to deal with a rapidly changing sector.

For more information, contact: Saul Guerrero, email: s.guerrero@actionagainsthunger.org.uk and Maureen Gallagher, email: mgallagher@actionagainsthunger.org

---

¹Horton et al (2010). Scaling Up Nutrition: What Will It Cost? (World Bank, Directions in Development/Human Development, 2010, p.19). This is based on the estimations of the authors that only 1 million of the total 19 million children suffering from SAM (c. 5%) are receiving treatment.

²Based on UNICEF’s more recent estimations (PD-Nutrition Section E-Bulletin, Issue 1, October 2012, p.2) the actual number of children receiving treatment is closer to 1,961,772 which suggest that the proportion of cases receiving treatment could be closer to 10%.

³ACF International, (2010). ACF International Strategy 2010-2015

⁴Needs based on SAM caseload as defined by National Nutrition Protocols

⁵In controlled settings, in uncomplicated incident cases with MUAC at or just below admission criteria/mild oedema.

⁶ALNAP (2006) Evaluating humanitarian action using OECDDAC criteria: an ALNAP guide for humanitarian agencies (Over-seas Development Institute, London, March 2006, p. 38-39)

⁷Adapted from Sadler, K, Myatt, M, Feleke, T and Collins, S (2007). A comparison of the programme coverage of two therapeutic feeding interventions implemented in neighbouring districts of Malawi (Public Health Nutrition, April 2007, 10(9), p.912)

⁸The SPHERE Project (2004). Humanitarian Charter and Minimum Standards in Disaster Response (First Edition, 2004, London, p.39)

⁹Vautier, F (1998). Selective Feeding Programmes in Wadjir: Some Reasons for Low Coverage and High Defaulter Rate (Field Exchange, Emergency Nutrition Network, Issue 5, p.17). Coverage calculated indirectly (using prevalence data against population estimates).

¹⁰All figures refer to point coverage.

¹¹As defined by National Nutrition Protocols and corresponding admission criteria.

¹²Doledec, David (2008). Impact of community mobilisation activities in Uganda (Field Exchange, Emergency Nutrition Network, Issue 34, October 2008, p. 15)

¹³Not all bottlenecks can be addressed rapidly or through community mobilisation. For a more detailed discussion, see Tanahashi, T (1978). Health service coverage and its evaluation (Bulletin of the World Health Organisation, 56 (2):295-303.

¹⁴Guerrero, S et.al (2010). Determinants of coverage in Community-based Therapeutic Care programmes: towards a joint quantitative and qualitative analysis (Disasters, Overseas Development Institute, April 2010, 34(2); 571-585)

¹⁵In countries where MUAC has not been incorporated as an admission criterion, such as Burkina Faso, rejection continues to be an important barrier to access.

¹⁶Schofield, L et.al (2010) SQUEAC in routine monitoring of CMAM programme coverage in Ethiopia (Field Exchange, Emergency Nutrition Network, April 2010, 38: p. 35)

¹⁷ACF West Africa Strategy, 2011-2015, p.6.

¹⁸CMAM Integration Guide draft, Feb 2011

¹⁹WHO (2010). Monitoring the Building Blocks of Health Systems: A Handbook of Indicators and their Measurement Strategies.

²⁰The first challenge relates to volunteer’s workload. Because of how important they are, many public health interventions incorporate the same cadre of volunteers into their outreach strategies. The result is an increasingly overburdened workforce capable of dedicating increasingly less time to each activity. The second challenge relates to the motivation of volunteers. The issue of motivation is linked to the issue of workloads and the issue of compensation. The tendency has been to motivate volunteers through the ad hoc provision of incentives, ranging from in-kind items (e.g. soap, sugar, t-shirts and bags) to cash payments. Whilst this commonly raises questions about sustainability, the bigger and more relevant issue is whether volunteers should be placed at the heart of community sensitisation and case-finding in the short, medium and long-term.

²¹Amthor R, Cole SM and Manary M (2009). The Use of Home-Based Therapy with Ready-to-Use Therapeutic Food to Treat Malnutrition in a Rural Area during a Food Crisis. J Am Diet Assoc. 2009;109:464-467

²²Ibid. p. 464

²³Linneman Z et.al. (2007). A large-scale operational study of home-based therapy with ready-to-use therapeutic food in childhood malnutrition in Malawi. Maternal and Child Nutrition (2007, 3 , pp. 206-215)

²⁴Proportion of affected population, based on programme admission criteria, receiving treatment.