Integration of CMAM into routine health services in Nepal

By Regine Kopplow

Regine is a former CMAM Advisor with Concern Nepal. She is a nutritionist with a background in rural development. She has worked in the field of nutrition with Concern in Afghanistan, Malawi, and Nepal and with UNICEF in Somalia.

The author would like to thank all the health workers and FCHVs of Bardiya district and the Concern Nepal team, for their contributions to the work reflected in this article. The author draws upon content of two reports authored by Saul Guerrero¹ and Lily Schofield².

This article describes a pilot project by Concern Worldwide in Bardiya district, Nepal to integrate CMAM into routine health services directly, without the more typical transition from an NGO led programme.

According to the Nepal Demographic Health Survey (NDHS), under-five mortality has nearly halved between 1996 and 2006, suggesting a good chance of meeting the Millenium Development Goal (MDG) 4 target of 34 deaths per 1000 live births in 2015 (Figure 1)³. However, acute malnutrition remains high - 12.6% global acute malnutrition (GAM), 2.6% severe acute malnutrition (SAM) (NDHS, 2006)⁴. Without addressing SAM in particular, under-five mortality in Nepal might remain above the MDG 4 target regardless of the remarkable achievements on mortality to date.

Participatory training in food groups conducted by the Concern intern Binita Sadaula in Bardiya district

In 2007, UNICEF carried out a feasibility study to explore the potential of the community based management of acute malnutrition (CMAM) approach to address SAM in Nepal. The study concluded that CMAM could be a useful tool but only if integrated into the routine health services provided through the Ministry of Health and Population (MOHP) and preferably linked to the integrated management of childhood illnesses (IMCI) approach which has already been introduced in Nepal. In order to reflect geographical and cultural differences of Nepal, piloting in the three agro-ecological zones (lowlands/terai, hills, mountains) was recommended. Furthermore local Ready to Use Therapeutic Food (RUTF) production should be explored to reduce costs, enhancing sustainability in the long run.

In 2008, Concern Worldwide signed an agreement with UNICEF for the CMAM pilot in Bardiya district located in the terai of the mid-west development region of Nepal bordering India. Two more pilot districts were identified located in the hills (Accham) and mountains (Mugu) in mid and far-west Nepal, covered by UNICEF with technical support from Action Contre la Faim (ACF) and later Concern (Accham only).

Pilot objectives

The primary objective of the CMAM pilot programme was to evaluate the feasibility of the CMAM approach in different districts and different agro-ecological zones in Nepal and to provide recommendations to MOHP in regard to treatment of malnutrition and the potential scale-up of the CMAM approach to most of the districts in the country. A secondary objective was to build capacity of local health structures, female community health volunteers (FCHV) and local partner non-governmental organisations (NGOs) to manage severe acute malnutrition (SAM) and to evaluate the effectiveness in increasing coverage of identification and effective treatment of SAM cases. While introducing CMAM to the health system, other interventions specific aspects were to be studied, such as costeffective ways to recognise and target most affected communities, health system capacity, women's/health worker's time allocation and the logistic/supply arrangements and management.

Pilot Strategy

MOHP, UNICEF, ACF and Concern developed a joint pilot strategy. A memorandum of understanding (MOU) outlined pilot details (objectives, strategy, roles and responsib- ilities):

1. All MOHP health facilities, health workers and female community health volunteers (FCHV) are to be involved in the CMAM activities reaching out to the entire population of the district (see Table 1).
2. The provision of CMAM services to be part of the daily routine of MOHP health workers and FCHV (without additional financial incentives) with technical support through the supporting international NGO.
3. For CMAM trainings, MOHP national and district trainers to be trained using a cascading down approach to train all MOHP health workers and FCHV of the district (see Table 2).
4. MOHP supply and reporting structures to be used and strengthened where needed.
5. Independent monitoring provided through field monitors of a local NGO, responsibility later to be incorporated into MOHP structures.
6. The pilot will not include a Supplementary Feeding Programme (SFP). This component will only be added in the case of severe food insecurity since the main causes for acute malnutrition in Nepal are currently related to inadequate child feeding and care practices, hygiene and sanitation, and health care utilisation.

Street drama about malnutrition and CMAM performed by the local partner NGO, Suryoday Sanskriti Pratisthan, in Bardiya district

National CMAM pilot protocols were developed jointly (Table 2) and training materials prepared for each level of training (manual for national and district level training of trainers, SC, OTP, FCHV trainer, and a pictorial flip chart for the mostly illiterate FCHV)⁶.

Implementation

In May 2009, the first SAM cases were treated by MOHP health workers in Bardiya district using CMAM treatment protocols. Children seen during routine health check ups are nutritionally screened (Mid upper arm circumference (MUAC)/oedema/WHZ) and admitted to the OTP or referred to the SC if they present with admission criteria (Table 3). Children return for their bi-weekly OTP follow up visits on the day of the week most convenient for the mother. In this way, acute malnutrition is treated in the same way as any other childhood illness - it is diagnosed through normal consultation and treated through regular CMAM services at the health post until resolved.

During the set up phase, field monitors of the local partner NGO, Community Development Organisation (CDO), provided technical support to MOHP health workers at CMAM outpatient service providing facilities (two days per week), and SHP and FCHV (on the remaining days of the week). Supervision is used technically to support and monitor CMAM providers. Weak areas are jointly identified and solutions developed. A supervision checklist guides both through the supervision visit and allows documentation of the observed performance. Checklists are analysed promptly and form the basis for a quarterly award system to acknowledge the best performing PHC/HP (OTP), best SHP and best FCHV. This creates healthy competition among CMAM service providers in order to motivate for better performance. Exchange visits between stronger and weaker health facilities support learning from peers rather than relying on external support.

With this support system in place, health workers and FCHVs felt confident and took full responsibility for CMAM services from the beginning. At no time during the project did Concern staff screen, refer or treat CMAM children. No additional MOHP staff were recruited or financial incentives provided for screening and treatment activities.

Table 4 summarises key activities carried out during the pilot project. The majority of tasks were implemented according to plan except the set up of the SC (planned for January 2009) and the implementation of community mobilisation activities (planned for the first quarter 2009). The MOU signed was not specific enough in regard to the roles and responsibilities of setting up CMAM inpatient treatment capacity (SC) for complicated SAM cases. No Concern CMAM officer was specifically assigned for planning and carrying out community mobilisation, thereby delaying activities significantly.

Admissions and impact of WHO Growth Standards

Prior to the start of the project, the expected case load was calculated using data from the Bardiya nutrition baseline survey (May 2008). At this planning stage, referral and admission criteria had not been finalised. Table 5 shows case load calculations⁸ using different SAM identification criteria. In line with the WHO/UNICEF joint statement⁹ the expected caseload in Bardiya district increases approximately threefold when moving to the new WHO growth standards from NCHS (Weight for height z score (WHZ)). A MUAC cut off point of 115mm leads to a similar increase. Figure 2 shows the actual OTP admissions in Bardiya. Within the first eight months, 1,213 SAM cases were admitted, 90% of the annual target.

In Bardiya district, FCHVs are the main contact for nutritional screening and referral of SAM children using MUAC <115mm and oedema assessment during home visits, mother group meetings and community events. At PHC/HP, the screening and admission protocol is by MUAC, oedema and WHZ but due to high work loads and difficulties integrating weight and height measurements into the work flow, MUAC and oedema checks are the main criteria used. Project data (see Figure 3) shows that 78.2% of SAM children are meeting OTP admission criteria by MUAC<115mm (35.4% MUAC only, 42.8% meet MUAC and WHZ criteria). WHZ accounts for 21.8% of admitted children only. Analysing the project data with NCHS reference and MUAC of <110mm, only 37.9% of the children would have been identified by MUAC. Using WHZ (NCHS) and MUAC >110mm, 54% (n=331) of children admitted in the Bardiya OTPs would have been classified as not severely acutely malnourished.

The increase of admissions by using a MUAC cut off point of 115mm has not resulted in the expected proportional increase of younger children¹⁰ being admitted in the OTP in Bardiya district (see Table 6).

There were early challenges, such as MUAC tapes with 110mm cut off point already in circulation by other programmes and unavailability of the new MUAC tapes until April 2009 making local procurement necessary. However, the introduction of the new SAM classification has simplified identification of SAM cases by FCHVs and health workers in Bardiya and has helped in more timely identification of cases.

Gender difference in admissions

It was observed that MUAC admissions into Bardiya OTPs for girls was much higher than for boys (boy: girl ratio 0.524) even though the nutrition survey had indicated equal SAM prevalence for both sexes using WHZ (WHO). In line with survey findings, WHZ admissions showed no difference between male and female admissions. It is assumed that severely malnourished girls have a higher risk of mortality than boys, resulting in a lower MUAC manifesting already in the very young. A first explanation for this gender disparity could be that admitted girls were found to be suffering more from general illnesses than boys (60.9% vs. 37.1%). Further research is required to confirm anecdotal explanations that gender biased child feeding and care practices were also a cause of these differences.

Performance indicators

Basudev Gautam, Nutrition Focal Person of Bardiya DHO, supervises practical exercise for FCHV checking nutritional oedema

Between May and December 2009, 878 children were discharged from the programme with an average length of stay (LOS) of 47 days¹¹ and an average weight gain (AWG) of 6.1g/kg body weight/day. This is in line with comparable CMAM programmes¹².

For the Bardiya pilot project, Sphere standards developed for resource-intensive, NGO-led emergency programming have to be used as a reference as standards for integrated, MOHP-led CMAM programmes are not yet available. Compared to Sphere standards, the CMAM pilot in Bardiya shows a low recovery rate due to many children defaulting with the majority of the defaulters (59%) missing their first follow up visit. The main reasons for defaulting were miscommunication between the caregiver and health worker, the mother's perceived recovery of the child and distance to the nearest OTPproviding health facility.

Gulariya HP in charge with CMAM field monitor from CDO jointly analysing OTP patient cards

A detailed analysis shows that 23% of defaulters had poor or no appetite when admitted to OTP. They would normally warrant inpatient care however, health workers admitted that caregivers often refuse a referral to the SC, leading to inappropriate outpatient treatment with unsatisfactory weight gain and caregivers misinterpreting lack of appetite as the child disliking RUTF.

With MUAC being the main CMAM entry point, the discharge criteria is adjusted to 15% weight gain and no other SAM criteria met (MUAC>=115mm, no oedema, WHZ>=-3SD). According to the pilot protocols, no minimum LOS is required. Figure 5 shows the corresponding WHZ (WHO) of 365 analysed cases when discharged as cured. For 62.5% of children discharged as cured, the 15% weight gain resulted in a WHZ >=-2SD. Also, 9.3% of children discharged as cured were found still to be severely malnourished (i.e. had a WHZ<-3SD. This was mainly due to lack of adherence to treatment protocols. Overall, health workers perceived a trigger point for discharge of 15% weight gain as a helpful tool to reduce the need for time consuming height measurements during OTP follow up visits. A minimum LOS could have prevented children from being discharged too early. However, a further increase in defaulters could result if the minimum LOS is not closely enough linked to the perceived nutrition status.

Coverage

A SLEAC (Simplified LQAS¹³ Evaluation of Access and Coverage) survey was carried out in November 2009. The coverage found across the district was classified as below the 50% target. Out of the 35 children who were missed out, 34% were eligible for OTP using WHZ only (MUAC<=115mm). This presented a challenge to the screening strategy which mainly relies on community based MUAC screening through FCHVs. Figure 6 summarizes the barriers to accessing OTP services.

The delay in community mobilisation activities is the main reason why more than 50% of caregivers of those children missed out and identified in the SLEAC survey were unaware of the programme. Due to a staffing oversight, activities to raise awareness about malnutrition and availability of CMAM services commenced only during the last quarter of 2009. A number of mobilization activities were subsequently implemented - street dramas, public cooking demonstrations, nutrition school days, orienting private practitioners/traditional healers and traditional birth attendances, conducting house-to-house screening in selected wards and broadcasting an educational radio programme and a nutrition song. Following this, admission numbers increased.

Integration

FCHV pictorial tally form for reporting screening activities

Concern's technical support was mainly focused at district level working towards integration of CMAM services into the existing MOHP health structure. The district level MOU, Concern's office within the district hospital compound and a joint CMAM bank account formed the basis for the collaboration with the district health office (DHO). The monthly meetings conducted by the DHO with all PHC/HP (district updates, Health Management Information System (HMIS) reporting) were/are used for identifying CMAM issues (monthly reporting, quarterly supply requests, general feedback). Although the implementation strategy was aimed at strengthening existing structures in order to secure the RUTF, Concern and UNICEF stepped in from the beginning to ensure RUTF availability at all times but without really testing the MOHP supply chain (Figure 7). Systematic drugs were supplied through MOHP without experiencing any shortfall.

With support of the field monitors, PHC/HP in charge prepared their monthly CMAM reports and submitted them along with their HMIS reports to the district health office. The compilation of monthly CMAM statistics for the district was gradually handed over by Concern and since early 2010 has been done by the DHO statistician. In January 2010, Concern reduced its presence in the district to one staff member only focusing on technical support to the DHO for the overall management of district nutrition interventions including CMAM. By the end of June 2010 Concern will have withdrawn from the district leaving CMAM in the hands of the DHO and his team.

Spider web prepared during a village development committee meeting in Bardiya district to collect feedback from CMAM users

The successful integration of CMAM into the daily routine of health workers and FCHV required a series of tiny adjustments. Contact points between FCHV and caregivers and their children suit CMAM screening and referral. Pictorial tally sheets to document screening and home visit activities were developed in a familiar design and incorporated into their already existing reporting booklet. At the health facility, children are registered in the IMCI registration book and medically checked following IMCI procedures. For the nutritional screening, a special form is designed linking IMCI with CMAM procedures and guiding health workers to making the right CMAM admission/referral decision. All systematic drugs prescribed at OTP and SC are on the MOHP drug list and provided free of charge. CMAM reports are compiled following the Nepali calendar and training manuals, reporting forms and patient cards are available in Nepali.

CMAM field monitor, Rang Bahadur Nepali, from CDO informing a community in Daudakala, Bardiya district about CMAM

Space and time was permitted for health workers implementing CMAM according to their individual resources and needs. Due to the short project period, integration successes are very much limited to areas where existing structures suit CMAM best (screening, referral, admission and treatment of SAM cases). CMAM components relying on weaker DHO structures (community mobilisation, monitoring) were subcontracted to local partner NGOs. Due to the limited time, the prevention of SAM through interventions aiming for behaviour change was not part of the CMAM pilot in Bardiya. However observations made indicate that government health services might have inadequate resources to provide individual nutrition counselling as part of IMCI, growth monitoring or CMAM.

Conclusions

CMAM Nepal Logo. Translation: Regularly screen the nutrition status of your child - receive nutrition counselling and timely treatment

The external evaluation comes to the conclusion that the screening, referral, admission and treatment of SAM children, performed exclusively by staff of the district health office (DHO), is done in line with pilot protocols and procedures even though no introductory or transition phase in which Concern or local partner staff carried out these activities was provided. However, the high defaulter rate and low coverage suggest that community mobilisation activities were not adequately prioritised. Greater integration at community and facility level requires further simplification of the CMAM protocols. It was the right decision to introduce the MUAC cut off point of 115mm and discharge based on 15% weight gain. However a well defined minimum LOS in programme is advised.

Finally, the delivery of RUTF through the MOHP supply chain is crucial with UNICEF/Concern reducing their involvement in this to technical guidance only. Delivery/logistics costs for RUTF have now been incorporated into the DHO budget for the coming financial year, pending approval by central government. Management of supply logistics through government remains a key challenge, more so than cost of supplies.

For more information, contact:
Regine Kopplow, email: Regine.Kopplow@gmx.de
and Ros Tamming, email: Ros.Tamming@concern.net

¹Final Evaluation of Concern Worldwide/MoHP Community-based Management of Acute Malnutrition (CMAM) Pilot Programme, Bardiya District, February 2nd - 19th, 2010, Nepal, Saul Guerrero.

²SLEAC Coverage assessment of CMAM pilot in Bardiya District, Mid-Western Region, Nepal, November, 2009, Report prepared by Lilly Schofield

³2010 report on Nepal's MDG progress prepared by the National Planning Commission (NPC).

⁴GAM: WHZ<-2SD and SAM: WHZ<-3SD. WHZ (NCHS) 2006 figures for same population are 12% GAM and 1% SAM.

⁵The FCHV structure has been introduced by MOHP throughout Nepal with one FCHV per ward functioning as a first consultation and referral point for basic maternal and child health care.

⁶Training manuals developed are based on: Training guide for community-based management of acute malnutrition (CMAM), November 2008, FANTA Project

⁷Referral and admission criteria are the same to avoid negative feedback through referred but later rejected cases.

⁸Total under five population Bardiya district (47,789) x SAM prevalence (2.8%) x assumed coverage (50%) x expected incidence (factor 2)

⁹WHO child growth standards and the identification of severe acute malnutrition in infants and children. A Joint Statement by the World Health Organization and the United Nations Children's Fund, 2009

¹⁰Joint Statement WHO/UNICEF 2009: "Using weightfor- height based on the WHO standards or MUAC less than 115 mm as admission criteria will select younger and less severely wasted beneficiaries compared to using the NCHS reference for weightfor- height or MUAC less than 110 mm."

¹¹Analysis of LOS and AWG for 401 children dischargedcured representing approximately one third of admissions in 2009

¹²Ethiopia (Wollo), project period 02/03 - 12/03, 590 cured cases, LOS 80 days, AWG 4-4.5g/kg/d; Malawi (Dowa), project period 01/05-12/05, 1,696 cured cases, LOS 45 days, AWG 5.8g/kg/d

¹³Lot Quality assurance Sampling