Nutrition in Emergencies

Assessment of Adult Undernutrition in Emergencies Report of an SCN Working Group on Emergencies special meeting, April 2001 Presentations of field data on adult undernutrition were made to the SCN Working Group on Emergencies in 1999 and 2000. As a result of the interest in these data and the increasing number of adult surveys received by the RNIS, the SCN published a supplement on the Assessment of Adult Undernutrition in Emergencies in July 2000. This led to a one-day special meeting on adult malnutrition held during the SCN 28th Session, the aim of which was to reach a common understanding of the recommendations on the assessment of malnutrition in adults and to identify practical steps to improve practice.

This meeting was convened to:

· reach a common understanding of the recommendations proposed in the supplement, and
· to identify practical steps to improve practice in the assessment of adult undernutrition.

Overview of the meeting

Presentations were made on the RNIS recommendations and operational agency field experience (Concern, CDC and Helpage), these were followed by an expert panel discussion¹. The participants then divided into small groups to consider the following topics and make interim recommendations for practice and research: (i) population based assessments, (ii) screening for selective feeding programmes, (iii) training, and (iv) operational issues. The following interim recommendations were agreed during the plenary.

Interim Recommendations for Operational Agencies

WHEN TO CONSIDER ASSESSING ADULT UNDERNUTRITION

The group felt very strongly that it is only appropriate to assess adult undernutrition in emergency situations in very specific circumstances. Assessments of population nutritional status should not routinely include adults because, in general, the nutritional status of the underfive population is a good proxy for the nutritional status of the wider community. However, it may be appropriate to consider assessing adult undernutrition in addition to children’s undernutrition in specific circumstances, for example:

· If the crude mortality rates begin to approximate or surpass the underfive mortality rates, suggesting that the population over-five is as vulnerable as the underfive population.

· If the prevalence of undernutrition is very high in the underfives and is not due to a health problem mainly affecting that age group

· If there is reasonable doubt that the nutritional status of children does not reflect the adult nutritional situation. For example in Bosnia or Kosovo it was suspected that older people were particularly vulnerable to undernutrition.

· If many adults attempt to enroll in selective feeding programmes or present to health posts

· If anecdotal reports of adult undernutrition are received

· If there is low coverage of food aid in dependant populations

· If data is required as an advocacy tool to lever resources

The group recommended that surveys of adults should not be undertaken unless the following pre-requisites have been met:

· A thorough contextual analysis of the situation has been undertaken including an analysis of the causes of undernutrition

· Technical expertise is available to ensure quality of data collection, adequate analysis and correct presentation and interpretation of results

· Clear and well-documented objectives of the adult survey are formulated

· The resource and/or opportunity costs of including adults in a survey have been considered.

Anthropometric assessment of adults should not be undertaken in isolation. An analysis of the causes of undernutrition should accompany anthropometric assessment. Data must also be gathered on other food security, nutrition, health, and economic variables. Nutrition information could include the results of assessments of children 6-59 months of age or other age and sex groups. Health data could include surveillance or survey data on the incidence of illness and death, especially illness and death due to those causes most closely associated with undernutrition, such as dysentery, measles, cholera, malaria, or others. Economic analysis may include market surveys and prices, migration, employment opportunities and household income.

In population-based surveys, valid sampling techniques, such as cluster or systematic random sampling, must be used in order to generalise the result. Crucial to this process is a clear definition of the population of interest and the objectives for the survey. For example, if adults are to be included in a survey of children less than five years of age and only households with young children are chosen, the sample of adults will not be representative of all adults in the population. Nonetheless, in most situations adults will be assessed along with children less than five years of age. In such cases, households should be the unit of sampling, and selection of households should continue until the desired number of both children and adults have been measured. All efforts should be made to prevent potential selection bias (i.e. inclusion of specific groups in the sample and exclusion of others) because adults are more often away from the home than children underfive, particularly healthy men.

Adult surveys should only include persons less than 18 years of age in special circumstances. Rapid changes in anthropometric measures due to pubertal development complicate anthropometric assessment of persons younger than 18 years. In contrast, older people should probably be included in assessments of adults where possible. Because older people may be more dispersed in the community as compared with adults and children, more houses must be included to ensure the sample size is adequate. In such surveys, agencies should consider using long-bone measurements such as arm-span or demi-span as a proxy for height in calculations of body mass index (BMI). In some situations, such as inflationary economies with many pensioners on fixed incomes, surveys may target only older people.

Because little agreement exists on the validity of proposed anthropometric indices for adults, adult surveys should aim to gather data on weight, height, sitting height, and mid upper arm circumference (MUAC). These data can be used to calculate BMI. Previous studies indicate substantial variation in BMI with Cormic index (sitting height divided by standing height: a measure of body shape). BMI should, therefore, be adjusted for Cormic index. Adjustment can substantially change the apparent prevalence of undernutrition in adults and may have important programmatic ramifications. MUAC measurements should always be taken. If immediate results are needed or resources are severely limited, surveys may include only MUAC measurements.

Because the interpretation of anthropometric results is complicated by the lack of validated functional outcome data and benchmarks for determining the meaning of the result, results must be interpreted along with the contextual information described above. The presentation of the results should include clear descriptions of all sampling and measurement methods. They should also present the prevalence of undernutrition as defined using multiple cutoff points. Such cut-off points should include the proportion of adults with BMI and Cormic-adjusted BMI below 16.0 and 17.0. Reports should also include a presentation of the distribution of values for all anthropometric measures and indices, for example cumulative frequency graphs or tables.

PRACTICAL RECOMMENDATIONS FOR ADMISSION TO SELECTIVE FEEDING PROGRAMMES

Before implementing selective feeding programs specifically tailored for adults, agencies should consider alternative strategies for improving household access to food. Where adult selective feeding programmes are necessary dry ration distributions are always more preferable to on-site wet feeding because of the opportunity cost for adults in attending these programmes. The main objective of therapeutic feeding programs should be the prevention of death.

THERAPEUTIC FEEDING

Where possible adult therapeutic centres should be integrated into wider programmes including TB treatment programmes, HIV support networks and other health programmes to allow referral of secondary cases of undernutrition.

Admission and discharge criteria should include a combination of anthropometric indices, social factors and clinical signs. Because little agreement exists on the validity of proposed anthropometric indices for admission and discharge, where possible weight, height and MUAC should be gathered for each individual. Individual agencies should decide on the indicator to use to determine entry and exit in each situation. This choice should take into account the known shortcomings of BMI, the lack of information on MUAC and the programme implications of their use (levels of need, resources available, presence of chronic or secondary undernutrition). Sustained weight gain should be an important element of the discharge criteria. Until the functional significance of anthropometric indicators is better understood, cut-off points determining when an individual should be admitted and discharged should be adapted to the resources available and the context. No generic set of social factors can be recommended for any emergency situation. These should be adapted to the particular situation (see the RNIS supplement for an example from South Sudan). Clinical signs should also be context specific. Some examples are: inability to stand and walk, bilateral oedema (Beattie grade 3 or more), sunken eyes. The relative importance of specific social or clinical signs is likely to vary and may need to be established locally.

The progress of adults admitted into centres should be monitored using: weight gain, presence of oedema, functional ability and MUAC. Individual patients should have a minimum weight gain of 5g/kg/day. As a reference, the duration of the individual’s stay should be between 15 - 40 days. Functional ability and MUAC should be recorded for research purposes. Failure to gain weight requires prompt investigation. Patients with secondary undernutrition (e.g. as a consequence of TB or AIDS) should be referred to more appropriate programmes.

RECOMMENDED RESEARCH PRIORITIES

The group recommended five areas of focus for operational research on adult undernutrition. First, proposed cutoffs for BMI, Cormic-adjusted BMI and MUAC should be validated using mortality and other data on functional outcomes, such as morbidity, ability to carry out routine activities, and reproductive health outcomes. Second, the variation in MUAC by age, sex and ethnic group is largely unknown. Data on MUAC from various groups, including both well nourished and undernourished adults, should be collected as soon as possible. Third, the aetiology of adult undernutrition, including adult nutritional oedema, requires further investigation. Health and food security indicators should be considered in this investigation. Fourth, anthropometric and functional methods to differentiate between acute and chronic undernutrition need to be developed. Fifth, anthropometric methods for assessment of undernutrition in the 18-25 year group, in older persons and in adolescents need further development. The SCN Working Group on Emergencies Thematic Group on Adult Malnutrition will further develop this research agenda and make steps towards its implementation.

Acknowledgements

The meeting and this report was possible through the support provided to the Food and Nutrition Technical Assistance (FANTA) Project by the Office of Health and Nutrition of the Bureau for Global Programs Field Support and Research at the U.S. Agency for International Development, under terms of Cooperative Agreement No. HRN-A-00-98-00046-00 awarded to the Academy for Educational Development (AED). The opinions expressed at the meeting are those of the participants and do not necessarily reflect the views of the U.S. Agency for International Development.

Special thanks go to the expert panel members for their invaluable contribution to the success of the meeting.

1 Expert panel members: Dr. Steve Collins, Dr. Bradley Woodruff, Dr. Carlos Navarro-Colorado, Dr. Arabella Duffield, Dr. Peter Salama and ...

ACC/SCN Refugee Nutrition Information System After a short break in production due to staffing changes, the SCN Secretariat is pleased to advise that a bumper RNIS report (issue # 32/33) was published in May. This report on the nutrition situation of refugees and displaced populations is available in hard copy from the SCN Secretariat (accscn@who.int) or can be downloaded from the ACC/SCN web site at http://acc.unsystem.org/scn/ RNIS # 34 will be published at the end of July 2001. We’d like to say HELLO to Tewodros Woldemariam who wrote recently from Derwonaji Refugee Camp in Eastern Ethiopia saying: By chance I found one of your early RNIS Reports (March 1998) in the archives of our store. I was really amazed by how much relevant information it contains. To be honest this is the only publication we have concerning activities in refugee camps. There is a total lack of information exchange on the general situation of refugees around the world. Rest assured Tewodros, we’ve added your name to our mailing list and from now on you will receive the RNIS Reports as they are produced!