Community-based Management of Severe Acute Malnutrition in BangladeshReducing Vulnerability to Malnutrition in Poor Cyclone-prone Communities

Project Team
Kate Sadler

Malnutrition is a major public health problem throughout the developing world and is an underlying factor in over 50% of the 10-11 million children under five years of age who die each year of preventable causes. In many areas where chronically high levels of acute malnutrition have been identified, there is a dearth of feasible strategies for identifying the condition and for delivering treatment within ongoing child survival programming. Improvements in care at health facilities are necessary, but in the poorest areas of the world primary health care facilities are often a long way from people’s homes and the opportunity and financial costs of seeking care for these families are high. This means that children from the poorest families are significantly less likely to be brought to health facilities, and may receive lower quality care once they arrive.

This study aims to test the effectiveness of adding the diagnosis and treatment of severe acute malnutrition (SAM) to the integrated management of childhood illness package, delivered by community health volunteers (CHVs) outside health facilities, against the standard of care for SAM in Bangladesh, which is referral from primary health care facilities for treatment in inpatient centers. The findings will be used to inform policy and practice for the diagnosis and treatment of SAM both in Bangladesh and across the developing world.

The principal investigator on this project has, over the last six years, been involved with the development of a new approach for the treatment of acute malnutrition in emergencies: community-based management of acute malnutrition (CMAM). This model of care is now used widely across Africa, with identification and treatment of children suffering from SAM delivered from primary health care facilities by primary health care practitioners. CMAM was recently endorsed by the WHO, UNICEF, and UNHCR.

However, with the problems that many poor families experience in accessing center-based health services, there have been calls to implement a household and community component of the management of childhood illness. This would serve the sick children who never reach facility-based services. Although such a component has been tested and is now being rolled out for conditions such as diarrhea and acute respiratory infections (ARIs), the community-based management of SAM remains facility-based with outpatient treatment delivered by, for the most part, trained health workers working out of these facilities.

This is a prospective cohort study that aims to examine the operational effectiveness of community case management (CCM) of SAM delivered by CHVs.

Community case management of SAM will be rolled out across 26 of the 61 unions in Bhola District, Barisal Division by the end of 2009. These unions (the intervention unions) have been selected by SCUS because they are deemed to have poorest health care coverage in the District. This is deemed to be a feasible roll-out strategy within the resources and capacities on the ground. Outcomes from the intervention unions will be compared with those unions in Bhola District that are not yet exposed to CCM of SAM (non-intervention unions) and therefore treat SAM using the standard of care.

Four hypotheses will be tested during this 17-month study:

  1. community case management of SAM will achieve acceptable clinical outcomes (recovery and mortality) when compared with international standards, and better outcomes than those reported by the standard of care for SAM;
  2. the cost-effectiveness of treating SAM by CCM is better than that of the standard of care;
  3. enabling community health workers to diagnose and treat SAM will increase utilization of SAM treatment services among children under two years compared to the provision of treatment through standard facility-based inpatient services;
  4. community health volunteers can treat children suffering from uncomplicated SAM as effectively as facility-based health personnel.

Data collection under this study drew to a close in 2011. This has included parallel-running study components that have examined the effectiveness of this treatment approach in terms of recovery from SAM and reduction of mortality, the level of coverage achieved, the quality of care delivered by CHVs, and the cost effectiveness of the program. In addition, considerable effort has gone into sensitizing the Bangladeshi health and nutrition community about the potential advantages of incorporating community-based management of SAM into ongoing services. As a result, key decision makers in Bangladesh now value the outputs of this study (see below).

By June 2010, community health volunteers had identified and treated over 700 children with severe acute malnutrition in one district of Barisal Division in southern Bangladesh. Over 92% of these children recovered and only one child died during treatment. A preliminary analysis of study data shows that average length of stay for the first 211 recovered children was around 32 days and average weight gain 7.9g/kg/day. These are excellent results and compare very well to international standards for therapeutic feeding programs (SPHERE, 2007) that stipulate a standard of greater than 75% for recovery. These data indicate that the quality of care delivered and acceptability of the program by the community was extremely good. Compared to other well-run similar programs in Africa, these results are far superior. In Malawi, for example, recovery rates in the area of 75%, an average length of stay of 42 days, and average weight gains of 5.4g/kg/day are commonly recorded (Sadler, Kate. “Community-based Therapeutic Care: Treating Severe Acute Malnutrition in Sub-Saharan Africa.” PhD thesis, 2008.).

The coverage (i.e., number of children suffering from SAM reached by this program) was also extremely good at 89% (Confidence Interval: 78.0% – 95.9%). Again, this is well within the international SPHERE standard for coverage for therapeutic feeding programs (>50%) and is one of the highest rates of coverage ever recorded for this type of program.

It is well known that untreated SAM carries a very high risk of mortality and morbidity for children. The high recovery rate and program coverage seen in this study represents a significant reduction in this risk and will have averted a large number of child deaths in the study area. Importantly, this study demonstrates that identification and treatment of SAM can be successfully delivered outside the hospital setting in Bangladesh by a community cadre of workers who have received the minimum of health and nutrition training.

As a result of the mobilizing and partnership efforts of this study’s team, key health and nutrition decision makers are now committed to the urgent need to address SAM in Bangladesh and see the significant advantages of doing this through a decentralized community-based approach. The process of using this work to change policy and practice around the identification and treatment of SAM both in Bangladesh and regionally has already started. This will include:

  • A national results dissemination meeting in Fall 2010. This will be led by the study team, including the national institutions that participated as co-investigators on the study. These are: Prof. Fatima Parveen Chowdhury, Director of the Institute of Public Health and Nutrition in Bangladesh; Prof. Syed Zahid Hossain, Head of Pediatrics at Sher-E Bangla Medical College Hospital in Barisal; and Dr. Nazneen Anwar, Deputy Program Manager for the Director General of Health Services, Bangladesh. During this meeting, study results will be presented and disseminated, and working groups will begin to map out how such an approach might be incorporated into national policy.
  • Support to the national-level malnutrition working group for development of national guidelines on the community-based management of severe acute malnutrition.
  • Development of a research “Phase 2″ which will include examining the potential for an approach to address moderate acute malnutrition among children. This too is a significant problem in Bangladesh (over 10% of children under five suffer from moderate acute malnutrition nationally) and is the underlying cause of a large proportion of child mortality in the country.

We also envisage that this work will have wider impact on the design of basic child survival interventions supported by the international community, particularly UNICEF and WHO.

This project worked in collaboration with a large Save the Children US development assistance program in Southern Bangladesh. It provided the research setting for one PhD student from the Friedman School of Nutrition Science and Policy at Tufts University, who provided ongoing support for design, data collection, and analysis.