Global CMAM mapping in UNICEF supported countries

Summary of review¹

A recent review commissioned by UNICEF set out to develop a global map on the status of Communitybased Management of Acute Malnutrition (CMAM) with a focus on severe acute malnutrition (SAM) in UNICEF supported countries, at policy and programming levels. Management of SAM has been strongly supported by UNICEF in the last decade and remains an institutional priority. In recent years there has been a rapid increase in the number of countries implementing a community approach to managing acute malnutrition, as well as the expansion of these services within country. However, as yet there is a lack of strong, up-to-date, reliable global, regional and country data to monitor the quality and scale of programmes and services. In order to address this, UNICEF commissioned Valid International to carry out a global mapping of Community-based Management of SAM programming and recommend ways to improve data availability and reliability.

Over the past eight years, the communitybased approach for the management of SAM without medical complications has become a key intervention for the management of acute malnutrition, gaining United Nations (UN) endorsement through a Joint Statement in 2007². There has also been a large shift from using management of SAM as an emergency intervention, to a service that is part of routine child health activities. In line with its remit for improving child health, and also as the Nutrition Cluster lead³, UNICEF is one of the principal organisations supporting the implementation and scale up of the CMAM approach with respect to managing SAM. UNICEF is the main provider of Ready to Use Therapeutic Food (RUTF), therapeutic milk (F- 75, F100) and ReSoMal that are an essential part of treating SAM. UNICEF also provides technical support and capacity building, both directly to Ministries of Health (MoHs) and via non-governmental organisations (NGOs), to better manage SAM. Central to UNICEF's work is monitoring and evaluation (M&E) and information collection to demonstrate impact of programming.

Methodology

A questionnaire⁴ based on the World Health Organisation (WHO) health systems framework was designed by Valid International in collaboration with UNICEF headquarters, Regional Offices (ROs) and Supplies Division (SD). Questions included qualitative information (general CMAM programme background/context, policy, financing and coordination, training/capacity development, drugs and therapeutic supplies) and quantitative information (caseloads, prevalence, access and coverage, performance indicators). The questionnaire was sent out in June 2010 to 77 UNICEF County Offices (COs), selected due on the basis of previous orders for therapeutic supplies or indicated by the ROs.

Sixty-nine UNICEF COs responded (90% response rate). Questionnaires completed by 55 countries^5,6 with active CMAM programmes were analysed in detail. A summary of the key findings are given below.

Achievements to date

The first Community-based Management of SAM programmes started in Sudan, Malawi and Ethiopia between 2000 and 2003, reaching 9 countries (with the addition of Uganda, Zambia, Haiti, Niger, Bangladesh) by 2005, with a total of 55 countries by mid 2010 (see Figure 1). A further 7 countries are planning to introduce this approach for 2010/11, as well as Zanzibar as an extension of mainland Tanzania.

In nearly half of the countries (n=26), CMAM was initially introduced as an emergency response to environmental (drought, flood, cyclone) or political (insecurity) crises and implemented with support from NGOs and UNICEF. The high prevalence of acute malnutrition alone was the determining factor for 17 countries. HIV was stated as being the direct reason to introduce Community-based Management of SAM for Mozambique, Zimbabwe and, to some degree, Cameroon.

Scale up

Regarding scale up, the findings show that the roll out of Community-based Management of SAM is proceeding in an informal and localised way, rather than being directed by a strategic national plan. This is partly the result of the early stage of adoption of the Community-based Management of SAM approach (many countries are at pilot stage) and its initial introduction often being in an emergency context and carried out by individual NGOs. The main constraints affecting CMAM scale up reported were:

1. Financial constraints to purchase RUTF
2. Government priorities and policies regarding Community-based Management of SAM, including lack of acceptance of importation of RUTF
3. Inadequate quality of existing activities e.g. some countries stated the need for programme reviews before deciding whether to scale up.

Technical support

External technical support was provided to countries for initial implementation, training and guideline development through consultant services or institutional agreements with donor funded technical organisations, such as Valid International and FANTA-2. This was often facilitated by the ROs and by international NGOs.

Caseload

The caseload is significant. During 2009, over 1 million children were admitted for treatment of SAM. The majority of children admitted to community-based therapeutic programmes were in East/Southern and Central/Western Africa.

National operational guidelines

To date, 52 countries (95%) have final or interim national operational guidelines or protocols for Community-based Management of SAM. Thirtysix of the 52 countries have had guidelines in place since 2009⁷. The existence of endorsed operational guidelines demonstrates acceptance from governments and MoH for the community based approach but does not automatically equate with its use and application by health centre staff.

Use of WHO Growth Standards

It was reported that WHO Growth Standards (GS) were now in use in the majority of countries (n=36). Although adoption of the standards is increasing, it has been a gradual process. In seven countries it has only been partial, i.e. not all measures are applied across the whole country. In a further five countries where the GSs are included in the national protocol, they are not being used. Other countries are not yet moving to the new GSs due to resource implications that will result in higher SAM caseloads.

RUTF supplies

UNICEF provides at least 80% of the RUTF supplies in 70% countries (n=37) and provides 100% of RUTF requirements in 43% of countries (n=23)⁸. The Clinton Health Access Initiative (CHAI) is a major donor for supplies in several countries (100% in Botswana, Namibia, Swaziland and 99% in Mozambique) and UNITAID is the key donor in Zambia. Local production of RUTF is increasing, currently occurring in six countries (Democratic Republic of the Congo (DRC), Ethiopia, Madagascar, Malawi, Mozambique, Niger) and is planned in another 10 countries. According to the UNICEF Supplies Division, global suppliers exist in France, USA, Dominican Republic, South Africa, Kenya and India.

Acquisition of RUTF, the main commodity enabling community-based treatment, is well documented, showing an impressive increase in tonnage purchased by UNICEF during the last decade (see Figure 2).

Programme review status

Nearly half (47%) of countries (21) have had a programme review carried out within the last 3 years. Five countries considered that it was still too early to conduct an evaluation (Guatemala, Haiti, Botswana, Zambia, Zimbabwe), and a further two countries had a review planned for 2011 (Ethiopia, Burundi). While many reviews were of pilots or localised NGO programmes, encouragingly there is an increase in reviews of national level programming and policy.

Primary health integration

Progress has been made in at least 50% of countries to integrate CMAM with other primary health activities including Integrated Management of Childhood Illness (IMCI), Infant and Young Child Feeding (IYCF) and HIV/AIDS programmes.

Integration with both IMCI and IYCF policies has occurred in 13 countries, with a further 12 noting integration with IMCI alone and five countries with IYCF. For IMCI integration, the addition of mid-upper arm circumference (MUAC) to community-IMCI screening was the main change made. However, for both IMCI and IYCF it was frequently stressed that even when included as part of key policy documents and strategies, roll out or the level of activity at health centre level was still limited.

With regard to HIV and tuberculosis, Ethiopia and Mozambique reported integration with HIV policies. Other country offices did not specify, although it is likely that many have links given the high HIV prevalence in many of the countries managing acute malnutrition.

Integration with national policies has happened in over half of countries. Results indicated that in 34 countries, a national nutrition policy exists in which Community-based Management SAM is integrated (finalised or draft). A further 10 countries do not have a national nutrition policy, while another 10 have a national nutrition policy but it does not yet include Community-based Management of SAM.

Challenges

Reporting systems

Despite the wealth of valuable information provided by UNICEF country offices, there are considerable information gaps and constraints in the overall CMAM data collection system. One of the most relevant constraints is the diversity of reporting systems with varying levels of complexity used in different countries, limiting comparability and compromising reliability of any analysis undertaken. This lack of consistent information may also be due to the lack of a nutritionist or a person in charge of data management and information systems in-country, especially in an emergency.

National guidelines

The review of national guidelines from 28 countries showed differences in terminology used and lack of standardisation in admission and discharge criteria, which can be expected given there is no standard international template to facilitate consistency. Of more concern, guidelines appear to be becoming longer and more complicated, especially certain protocols and reporting requirements. This will affect the understanding of frontline health workers who have to interpret and use these.

Service provision and coverage

There is confusion between differentiating between the geographical distribution (service provision) and coverage (% of SAM cases treated).

It is difficult to draw conclusions at this stage as to what extent services are currently provided globally. Data on service provision was informed by just over half the countries (n=29). These indicated that, for the most part (18 countries), Community-based Management of SAM has, as yet, only been adopted in a small minority of administrative areas. In terms of the percentage of health facilities offering these services (clearly stated by 22 countries), the geographical distribution ranged from 1.4% to 94%. Half of the countries (50%, n=11) have Community-based Management of SAM available in less than 30% of existing facilities. It is important to note that this does not give an accurate picture of how programming is progressing without knowing whether the country is aiming for countrywide roll out or to support 'hotspots' with high rates of malnutrition.

Accurate estimates for programme coverage proved difficult to obtain due to the diversity of methods used, inaccurate information on population, SAM prevalence and incomplete caseload data.

Targeted coverage surveys using specific methodology e.g. CSAS⁹ or SQUEAC¹⁰ were specified by very few countries, although it is known they are being increasingly used with support from UNICEF and NGOs.

Information System: data collection and management

The lack of systematic collection of information from programmes represents a limitation to monitoring needs, gaps and impact. Only 23 (42%) countries have a database to manage and monitor the CMAM programme. Only a few countries mentioned a system to improve the chain of actions involved in information collection, including increasing the number of reports received at national level.

The responsibility for analysing the information was varied, but irrespective of where the responsibility falls, the findings clearly suggest that systems for the collection and transmission of reliable data need to be strengthened. The National Nutrition Centre/Section was responsible in the majority of countries (n=18), but also the M&E/ Health Information System (HIS) departments in eight countries, nutrition and statistical departments together in two countries and a further seven countries listed just the MoH or a specific section including family health or health promotion.

The transfer of information from local level to national level was sometimes outlined as a lengthy and often weak process. It ranged from direct transmission from health centre to national level to transfer of data in a series of 2, 3 or 4 steps, depending on the system and geographical divisions within country. Some countries rely on NGOs to transmit information and UNICEF still plays a significant role in facilitating data transfer, while others have focal points responsible for data transmission.

Feedback and action resulting from information supplied by countries is needed, as data collected is often not analysed or acted upon.

SAM caseload

Just over one-fifth of countries (22%, 12/53) could not provide caseload data for 2009, either because no data collection system exists or because information is held by MoH or implementing partners. Only 53% (29/55 countries) could provide any caseload data for January-May 2010.

Country caseloads are not directly comparable given the different stages of implementation and context influencing the numbers being admitted. For example, caseloads for 2009 ranged from small pilots with low caseloads e.g. Indonesia (n=73) and Timor Leste (n=230) to extensive countrywide programming of around 125,000 to 130,000 for Ethiopia and 124,187 in Niger in 2009.

Key performance indicators

Key performance indicators, such as the proportion of children who recovered, died or defaulted, were not provided by 26% (14/53) of countries for 2009. For these countries, there was either no system in place or the information was not available to UNICEF. In addition, 33% of the 39 responding countries received <50% of monthly reports at national level.

Of the 39 countries providing data, performance indicators for 22 countries met all Sphere standards, while 17 had one or more indicators that did not. Strong and weaker results were found across all regions. It is not advisable to compare these data further due to the fact that many country performance indictors did not add up to 100%, some countries included non-recovered/non response in calculations and some did not, and the overall number of exits is very variable.

The reliability of performance indicators is questionable. The evidence suggests that the figures given for performance indicators themselves must be questioned. Only 39 countries out of 55 were able to provide information with respect to the percentage of monthly reports received at national level and 33% (13) of the 39 countries had received less than 50% of reports at national level. Reports received may well have come from stronger or better supported programme sites, thus biasing the overall findings. It was also found that the monthly reports were often received late or on an erratic basis. Only three countries received 100% of reports.

HMIS/HIS

While many countries recommended including SAM indictors in the Health Management Information System (HMIS), only 14 (25%) have started this process with one or more indicator included. Ten countries emphasised the importance of HMIS integration, merging key indicators with existing HIS/HMIS systems where possible, or as a subsection within HMIS. The number of indicators included varied from just the number of SAM cases admitted, to several indicators, including caseload and performance. If multiple indicators were included, an additional sheet was sometimes added to the standard HIS form. Caution is given against having too many indicators included if the system is to be maintained accurately.

Funding

The reliance on short term or emergency funding for therapeutic supplies, capacity building and MoH support activities was the major reason given for delaying and disrupting further scale up of activities and services. Information from the questionnaire demonstrates that short term funding has been available for starting Community-based Management of SAM in many countries, particularly where this has been introduced as part of an emergency response. Longer term funding is an important consideration as this intervention is gradually integrated into the routine activities of government health structures and moves beyond a purely short term emergency intervention.

UNICEF is the major provider of SAM treatment supplies and/or activities in all countries. Nutrition therapeutic supplies are also acquired by national governments, NGOs, international agencies and foundations. Multiple donors are the norm in virtually all countries. The most prominent in terms of supplies are UNICEF, the Clinton Health Access Initiative (CHAI)¹¹, USAID, ECHO, and the Government of Japan. Similarly, the most frequent donors for activities are UNICEF, USAID, ECHO, WHO, WFP and the Government of Japan. Although these provide the bulk of funding for this approach, the support of other donors and individual NGOs including Medecins sans Frontieres, Concern Worldwide, Action Contre la Faim and Save the Children are significant in financing activities and to some extent supplies and distribution in a number of countries.

The need for long term funding for RUTF and capacity building components of Communitybased Management of SAM was raised by several UNICEF CO staff and highlighted in the training considerations (below).

Training

Training strategies, methodology and course length, in-service and pre-service training varied greatly between countries. As yet, only half the countries (n=28) have a country training strategy. This reflects the fact that the focus of Community-based Management of SAM has to date been largely localised. Although expansion of sites is now proceeding and the number of trained personnel is increasing, the design and implementation of a coherent national strategy is only starting to be realised. In-service training was used in the majority of countries (n=30) but pre-service training as yet is only taking place in nine countries.

The information provided highlighted that there is no standardised training model in use across countries. Time dedicated to Communitybased Management of SAM training ranged from 1 or 2 days to 11 or 13 days. The duration of community volunteer training varied from 2 to 5 days.

Recommendations to improve global mapping of CMAM

1. Develop a global SAM reporting system

The development of a well-structured, reliable information system at each level of the information chain is fundamental to understanding the current situation and to appropriately directing resources to improve the quality of programming. The minimum information to be collated, analysed, and interpreted at HQ level should form the basis of a global SAM reporting system that is useful for each level. Where possible, this should also be accessible to external stakeholders, such as on a password basis. At a later stage, this could be extended to include IYCF and surveillance data (e.g. collating nutritional surveys and connecting to the Integrated Phase Classification (IPC), Health and Nutrition Tracking System (HNTS) and cluster). In the first phase of development, it is recommended to focus on management of SAM only. Specific recommendations include the following:

Country level recommendations

It is recommended to simplify and harmonise the reporting system (with MoH) for management of SAM programmes. This would involve using standardised key indicators and information between countries, if possible standard reporting templates, and, importantly, use of the same reporting system within country by all implementing partners. This includes greater linkage of reporting between HIV and SAM information systems to avoid double counting or omitting caseloads covered by HIV support. Only a few countries reported linkages between these services, although many countries have high HIV prevalence and extensive programming for HIV infected children and HIV-infected adults using RUTF.

The percentage and timeliness of monthly reports received at national level should be increased by making the stakeholders more accountable for the reporting system. Clarity on reporting obligations from UNICEF-HQ to country level will ensure countries place more focus on this issue.

Regional level recommendations

Additional technical or M&E support for COs should be provided from people experienced with CMAM. This could be a multidisciplinary team based in UNICEF ROs or may include experienced staff from other technical agencies. This could help with issues including implementation, scale up and improving information flow, management and analysis of data.

HQ recommendations

Technical support: dedicate technical or M&E staff time to establish, manage and update the global SAM reporting system, in coordination and collaboration with the Supply Division and linked with ROs and COs. Ideally this will be a web based system.

2. Measuring service provision and coverage

It is necessary to develop a common way of indicating the service provision or geographical distribution of services available to treat SAM. It is recommended to calculate the number of facilities providing services to treat SAM out of the total number of health facilities. It is important to specify whether this information is provided for the whole country or only those parts of the country that are targeted for Community-based Management of SAM programming. Both pieces of information are important as they demonstrate different aspects of service access.

Increased use of localised coverage assessments will encourage reporting and understanding of coverage and barriers to accessing services.

3. Develop indicators for measuring integration into health systems

SAM indicators should be integrated into HIS/HMIS and agreement reached on which indicators can be integrated into HMIS. This should use experience from countries that have started this process and liaison with WHO-HQ.

Integration indicators/information should be developed that demonstrate progress into health initiatives including IMCI, HIV and tuberculosis, incorporation into annual health plans, health financing, pre-service training for medical staff etc. A national budget allocated to SAM could be useful.

More resources for M&E staffing, logistics/transport, training of supervisors, and information system focal points are required for the integration of CMAM into the health system to become a reality. Improving capacity and MoH resources are part of the integration process and should not be neglected or underestimated when calculating support or scale up needs from UNICEF.

4. Strengthen capacity development

As CMAM continues to develop and evolve, it is important to ensure technical staff, implementing partners and key stakeholders are kept updated with new evidence (including cost and integration), guidance and resources to develop country programming appropriately. Recommendations include:

• Technical support more readily available for personnel involved in nutrition programming, data information systems and M&E regarding information flow, management, analysis and supply forecasting.
• Include in the country programme plan further technical and logistical support to appropriate MoH staff to improve quality and timeliness of reporting, given data are often collected and databases maintained by MoH personnel.
• Update, develop and disseminate the Programme Guidance for Community-based Management of SAM (an internal UNICEF document).
• Promote and facilitate CMAM forum and technical discussions to improve quality of information and mapping. Share successes and strategies with countries that are seeking guidance on start up or scale up.

Conclusion

This exercise has provided a good start to mapping Community-based Management of SAM, providing a large amount of information in a number of areas. However the report emphasises that:

• Caution should be taken when making country comparisons, given the wide range of contexts (e.g. emergency and development, high and low levels of acute malnutrition, secure or insecure), different phases (e.g. time since start up, presence of external support, operational research) and different objectives/priorities (countrywide vs localised response only).
• Wider input from MoH and implementing partners is needed to both validate and expand on the current information held for each country.
• CMAM is dynamic, both in terms of the approach and developments in the management of SAM and also in the number of countries and areas where this is being implemented.
• While information is needed for quality and accountability, it is essential to keep any information collection as simple and practical as possible and to only collect minimum information that is useful at all levels i.e. country, region and global.

UNICEF has already started to address some of these issues. For more information, contact Ilka Esquivel, Senior Advisor Nutrition Security/Emergencies, email: iesquivel@unicef.org

---

¹Global Mapping Review of COMMUNITY-BASED MANAGEMENT OF ACUTE MALNUTRITION with a focus on SEVERE ACUTE MALNUTRITION. Global Mapping Review of Community-based management of acute malnutrition with a focus on severe acute malnutrition. Nutrition Section, Nutrition in Emergency Unit, UNICEF NY and Valid International. March 2011

²Community-based Management of Severe Acute Malnutrition A Joint Statement by WHO, UNSSCN, UNICEF May 2007

³Tracking progress on child and maternal nutrition: a survival and development priority. UNICEF, Nov 2009

⁴For any additional documents pertaining to the review, please contact UNICEF New York Nutrition in Emergencies office.

⁵Questionnaire response: 55 'countries' are used as the denominator for analysis, since during this review, North and South Sudan were the same country, They are counted in this report as separate programmes. UNICEF CO staff members are the source of information for the data compiled. Where possible clarifications and precisions were made with relevant staff but it was not possible to cross check with other CMAM information.

⁶UNICEF CO staff members are the source of information for the data compiled. Where possible clarifications and precisions were made with relevant staff but it was not possible to cross check with other CMAM information.

⁷In Bolivia, the guidelines referenced (2002) predate the introduction of CMAM in 2006.

⁸It was commented that the UNICEF Supplies Division fore cast only includes supplies provided, and needs met, by UNICEF, thus not giving the full picture as often countries receive supplies from various other sources.

⁹CSAS: Centric Systematic Area Sample

¹⁰SQUEAC: Semi-quantitative Evaluation of Access and Coverage

¹¹CHAI was previously known as the Clinton Foundation