World Vision programme for severe acute malnutrition in SNNPR

By Dr. Sisay Sinamo and Dr. Gedion Tefera

Dr Sisay Sinamo is Coordinator for the Health and Nutrition Coordination Unit with World Vision Ethiopia. A medical graduate from Addis Ababa University, he holds a Masters in Public Health. He has worked in areas of nutrition for about 10 years with practical field experience in community based nutrition, PD Hearth programming, emergency health and nutrition and SMART in a number of countries.

Dr Gedion Tefera is emergency health and nutrition programme manager with World Vision Ethiopia. He is a medical graduate from Jimma University, Ethiopia and holds a Masters in Public Health (MPH). For the past 4 years, he has worked in emergency nutrition and community based nutrition projects like essential nutrition package and PD/hearth.

We would like to express our thankfulness to the community where World Vision works and to our partners, especially the Ministry of Health. We also would like to thank World Vision Ethiopia for allowing us to publish our field generated learning. Special thanks go to World Vision Canada/Nutrition Centre of Expertise staff who welcomed this opportunity to publish, and to the ENN staff who supported us during field level documentation and editing of this article.

Follow up appetite testing at a health facility

World Vision Ethiopia (WVE) is a non-governmental organisation (NGO) that has been working in Ethiopia implementing crosssectoral relief, rehabilitation and development programmes since 1971. A WV supported programme to support community based management of acute malnutrition¹ in the Southern Nations and Nationalities Region (SNNPR) has been in operation since November 2006.

WVE implements cross-sectoral programming through Area based Development Programmes (ADPs). These operate independently of each other while benefiting from regional programme office oversight and national level coordination. The 'life-cycle' of an ADP is usually approximately 15 years with a programme planning review following a 5 yearly cycle (current Durame ADP plan is for 2006-2010). Programming partners and stakeholders include multi-sectoral faith-based organisations (FBOs) and community-based organisations (CBOs), as well as other international NGOs (INGOs), NGOs, communities and the Government of Ethiopia (GoE).

Context

The SNNPR region is characterised by a highly localised pattern of micro-climates. With 45% of the region's Gross Domestic Product (GDP) relying on agriculture, there is a strong correlation between livelihood and seasonality for individuals and communities alike.

A nutritional survey conducted by WVE in June 2006 in the Durame ADP revealed a global acute malnutrition (GAM) rate of 8.5% and severe acute malnutrition (SAM) rate of 1.3%. The crude mortality rate (CMR) and under 5 mortality rate (U5MR) were 0.31 and 0.89 deaths/10,000/day, respectively. A further WVE rapid assessment using mid upper arm circumference (MUAC) in September 2006 found substantial moderate acute malnutrition (n=2207) and severe acute malnutrition (n=258) in children less than five years of age. Based on these assessments and at the request of the regional and zonal health offices, it was decided that a programme to manage acute malnutrition at community level should be implemented by WVE from October 2006. Accordingly, a Memorandum of Understanding (MoU) was agreed between stakeholders for the programme to run for one year until September 2007. The programme was supported by Valid International and implemented by Ministry of Health (MoH) staff and three WVE staff hired to coordinate the activities.

Programming approach

Commitment was obtained from the health bureaus at different levels and UNICEF agreed to provide Ready to Use Therapeutic Food (RUTF) and IMCI² drugs. An understanding was reached to establish a Stabilisation Centre (SC) at Wota health centre. A developmental approach was adopted towards programming, to develop capacity of the MoH to provide effective treatment for SAM within the routine health system. Orientation for MoH staff took place between mid November and early December 2006. Eight MoH staff from seven OTPs and one health staff from the SC were trained on OTP and SC case management. An additional 19 WV staff from Ethiopia, Kenya, Sudan, Somalia, Uganda, Canada, US and World Vision International (WVI) were also trained.

Admissions started at seven OTP sites and the SC. A Supplementary Feeding Programme (SFP) component was not included, as this should be provided by the GoE Extended Outreach Strategy/Targeted Supplementary Feeding (EOS/ TSF) programme.

Key lessons from the 2008 programme

The end of term evaluation of the programme identified a number of key lessons:

An effective linkage between the OTP and EOS/TSF programmes was not realised during the programme. Due to the timing of screening and registration for the EOS/TSF programme (every six months), children discharged from the OTP were not able to receive a timely supplementary food ration from the programme. Hence, on discharge, children were provided with a protection supplementary food ration until the next screening and registration.

The community outreach activities are conducted primarily through the efforts of volunteer community health promoters (CHPs) and supervised by the health extension workers (HEWs). The CHPs are responsible for referrals from the community, making them the gatekeepers to the OTP programme. However they are not supervised effectively when conducting anthropometric measurements. In addition, the role of the CHP in the community does not include frequent and comprehensive screening for malnutrition. Unless other community screening measures are employed, the OTP service coverage is likely to be low, possibly resulting in late presentation and increased risk of mortality. Future community sensitisation and mobilisation efforts should aim to disseminate knowledge of community based management of acute malnutrition more widely through other channels, to encourage self referral.

The outcome indicators for the programme all exceeded SPHERE standards (see Table 1).These outcomes also compared very favourably with other community therapeutic care (CTC) programmes. The period coverage of the programme was estimated at 67% which, multiplied by the percentage of children cured (95%), suggests approximately 64% of needs were met. Rates of reported weight gain in the programme were lower than expected for a CTC programme (5g/kg/day). This was likely due to the RUTF ration being shared during times of need. Slow weight gains during the programme were not reliably followed up by clinicians according to the 'alert protocols' and discharge decisions were inconsistently applied. These factors combined to lead to an extended stay in the programme for some individuals, which meant an increased use of resources (RUTF and workload) and ultimately reduced the cost effectiveness of the intervention.

The weak supervision by clinicians of the MUAC measurement taken to admit children meant that, in many cases, it was effectively the CHP making the decision to admit the child to the OTP. Training of staff was frequently weak with regard to follow up and management of cases. A weakness in logistical management at all levels was found. The main weakness was in ordering and regulation of the RUTF supply.

As the programme evolved, the admissions to the SC and OTP decreased over time (see Figures 1 and 2). This is indicative of an increasing skill level amongst the OTP clinicians in cases management, as well as of the beneficial effect of the decentralised public health approach of community based management of acute malnutrition, i.e. early presentation and timeliness of treatment. A general improvement in food security over the time period may also have accounted for some of this positive trend.

Durame ADP programming plans for 2008 included interventions that aimed to tackle underlying causes of child malnutrition and mortality. Some of the more relevant programmes included care and support of people living with HIV (PLWH) and orphans and vulnerable children (OVCs), household food security programmes, water and sanitation projects to provide potable water and income generation activity (IGA) schemes. These programmes could have been linked to the community-based management of acute malnutrition programme, by virtue of their shared objective to impact their livelihood and work with vulnerable caregivers and children.

Follow-on programme to manage SAM (2009-2010)

In 2009, Ethiopia was affected by another short term mid year rain failure. Shone, Durame, Qacha Birra and Omosheleko ADPs in SNNPR were amongst those woredas most severely affected and had not yet recovered from the effects of the 2008 food crisis.

The lessons learnt from the Durame project (November 2005-December 2007) were instrumental in designing a programme to manage acute malnutrition in the four ADPs. The programme began in October 2009 and was based upon Government national guidelines for the management of SAM at both facility and community level³. It was developed in response to findings of a Rapid Nutritional Assessment conducted in two ADPs (Omosheleko and Quachabira) in March 2009 and MUAC mass screening conducted by the EOS programmes in December 2008 in Shone and Durame ADPs. See Table 2 for the prevalence of GAM and SAM in the four ADPs. Additionally, early warning reports from the four ADPs indicated deterioration of the food security situation in the areas due to lack of early rain (March and April 2009). The government identified the four ADPs as high priority areas for nutrition response.

*Proxy data for GAM and SAM since based on rapid/mass screening.

In 2008, the GoE had already started implementing and scaling up SAM management in the four ADPs and integrating this within the routine health care delivery. However, the programme had many capacity gaps that needed to be filled. In 2009 there was no adequate rain during the short rainy season and the local administration and regional Emergency Nutrition Coordination Unit extended a request to WVE for support. Furthermore, MOH and UNICEF with other partners planned to establish one OTP per kebele and one or two therapeutic feeding unit (TFU) SCs per woreda. WV contributed to this initiative through a partners' capacity building approach and helped expand and strengthen the existing programmes in the four ADPs' operational areas.

The current project is being implemented for 12 months from October 2009. It targets an estimated 4,559 children under five years of age with severe acute malnutrition in four ADPs. A key aim of the project is to help build the capacity of government staff in the health centres, health posts and community volunteers for smooth integration of the emergency intervention into the government system. Areas of partnership are reflected in Table 3. Key elements of the project envisaged at the outset were:

• Implementation of community-based nutrition projects in parallel with the emergency intervention using the Essential Nutrition Actions (ENA) Approach. Community education to include infant and young child feeding and maternal nutrition practices.
• The project has a number of capacity building elements and will be handed over at the end of the funding period (September 2010). During transition, the MoH will start to run most of the project activities.
• Links will be established between this project and other ADPs' food security and development projects that impact household food production, consumption and income. The OTP and SC will be linked with the existing EOS/TSF with the 'back up' of a WV protection ration (see below). Acutely malnourished children will be identified through community-based screening or by self-referral.
• The ADPs will conduct a local capacity assessment on management of acute malnutrition in the community to identify gaps. Trainings will be designed based on this assessment. During the initial assessment conducted before the programme started, there are no operational SC/OTPs, facilities. Hence, one facility will serve as an SC and four or five OTP sites for initial opening in each of the woredas. These would subsequently be expanded and strengthened based on need. Where SCs and OTPs are already in operation, these would be strengthened based on assessment findings.
• WVE and woredas MOH will conduct a oneday orientation session for Kebele leaders, HEWs, CHPs, school teachers, and community workers already working on community mobilisation, active case detection and defaulter tracing. The HEWs from the most severely drought affected kebeles will be trained first, followed by those in less severely affected woredas.
• WVE will be responsible for the purchase of essential medicine and other supplies needed, as well as hiring of vehicles for transportation of supplies and referrals, assisting government health staff with screening and admission, periodic monitoring and supervision of the project implementation together with the woredas health offices. WVE will also print the different registration cards (SC charts, OTP admission and ration cards, etc) and monitor the recording and the weekly and monthly reporting together with the woredas' health offices.
• WVE will liaise with UNICEF and Regional Health Bureau for the timely provision of drugs and supplies to the health facilities through existing systems for logistics transportation. WVE will fill gaps in the logistics pipeline, if needed.

The key programme design changes that emerged from lessons learnt from the 2006/7 programme were:

• WV staff were not used to implement directly the programme but focused on capacity building of partners, e.g. clinical officers and HEWs, adopting a minimum support approach.
• WVE supplied a protection ration for children discharged from the OTP as it could not be assumed that these children would be enrolled on the EOS/TSF.
• The referral system to SCs was changed to rely on self-referral, where mothers/ caregivers now take their children from OTP to SC. Previously WV had hired vehicles to transport 'complicated' SAM children from OTP sites to SC.

Six month review

Local MOH have taken the lead in implementing the OTP, SC and community mobilisation components, as well as beneficiaries follow-up in the community and referral to the EOS/TSF. Volunteers have screened and referred children to the OTP sites.

After the first six months of the programme (to March 2010), there are already a number of achievements:

Health facility assessment and stakeholder analysis were conducted involving 127 health posts and twenty health centres. WVE conducted joint woreda level planning workshops based on these assessments. The capacity gaps identified were presented to local MOH office managers resulting in discussions and joint plans regarding how to fill these capacity gaps.

By the end of March 2010, the programme had scaled up and was being implemented in 125 kebeles in the four ADPs. A total of 761 volunteer CHPs have been trained and are working with the HEWs in OTP sites. A total of 113 OTP sites have been strengthened/established. By the end of February, 2010, 1116 children were in the OTP, with the figure rising to 1438 children at the end of March 2010.

Children discharged from OTP sites received CSB/Famix and vegetable oil. The ration comprises 8.3kg/month of Famix and 1kg of vegetable oil per child per month. A total of 52.016MT of CSB/Famix and 6.153MT of vegetable oil were distributed up until the end of March 2010.

Other activities completed during the first six months of the programme included renting cars to provide onsite technical support and supervision, printing and laminating necessary OTP guidelines and protocols, and hiring a nurse and two RUTF distributors in each ADP project area. In some ADPs, food was provided for caregivers while they were staying at the SC. Two planned activities that were not possible were purchase of RUTF buffer stocks due to limited budgets as a result of the economic crisis, and monthly review meetings with government health workers, due to competing work demands on government staff.

Major challenges of the programme noted at the six month review included:

• Delayed staff recruitment
• Lack of routine medicines in some OTP sites
• Irregular use of tally sheets at OTP sites
• Delay of purchase of some commodities, such as computers and essential medications, due to lengthy purchase processes.
• Predicted shortage of CSB/Famix and vegetable oil in the coming months.
• Limited capacity of staff to use the WV database

Despite these challenges, OTP programme outcomes are very good, reflected in Table 4 using Durame ADP as an example.

Minimum support: PD Hearth

WV has recently completed an exploratory PD Hearth programme in Atbi Womberta ADP where a SAM management project had been phased out to a 'minimum support project'.

A volunteer community health worker conducting screening

Among all children under 5 years measured in the district, a total of 240 children were found underweight (weight for age <-1SD) of which 51.7% were severely under weight (<-3SD) and 31.7% were moderately under weight (>-3 and <-2SD). Children found with oedema and severe wasting were referred to the health facilities. A 'Positive Deviant Inquiry' was conducted with the 'Positive Deviant Mother' to draw on her experiences of child care, feeding and health seeking behaviours and in order to develop the feeding menu and key health messages to promote during the Nutrition Education and Rehabilitation Sessions (NERS session).Twelve days of NERS sessions were organised, with each session composed of 10 non-PD mothers and two PD Mothers. For each group, a 12 day feeding session was conducted with close monitoring and with follow up by the PD mothers and the HEWs. Growth monitoring was conducted on admission, during follow up, on discharge and after 12 days discharge and home follow up. A total of 240 mothers and their children participated.

Over a 1 month period, nearly one quarter of the children (24%, n=58) gained 0 - 200g in weight, 13% (n=32) gained 200 - 400g, 28% (n=66) gained 400g - 1kg and nearly one third (30%, n=71) gained over 1 kg. Five children (2%) in the programme lost weight and eight children (3%) did not respond. Over half of the children (58%) were rehabilitated successfully based on PD hearth programme discharge criteria (weight gain of at least 400g). Weight for age z score was used to determine the outcome. The proportion of children with severe underweight declined from half (51.7%, n=124) to under onequarter (23%, n=55) of the children. Follow up assessment was conducted and 85% of the mothers had managed to teach their neighbours to practice the new skills they had gained.

Based on this and other experiences, WV consider that in resource poor settings, PD hearth is a practical and sustainable behavioural change approach to prevent malnutrition and contribute to child well-being outcomes. WVE emerging policy in Ethiopia is that when SAM continues at low levels, perhaps following an emergency intervention, OTP can be mainstreamed into the regular health system to provide services for acutely malnourished children with PD Hearth focusing on preventative rehabilitation services in the community.

For more information, contact: Sisay Sinamo, email: Sisay_Sinamo@wvi.org and Dr. Gedion Tefera, email: Gedion_TeferaW.Michael@wvi.org

¹Formerly called CTC (community based therapeutic care).

²Integrated Management of Childhood Illness

³Protocol for the management of SAM. Ethiopia-FMOH, 2007.