Name: National Micronutrient Survey
Origin: Developed by OMNI in conjunction with the Honduran Ministry of Health and International Eye Foundation (IEC) for use in Honduras (1996). Modified versions locally adapted were used in Morocco and Zambia (1997), and in Mozambique (1998).
Source: OMNI
Basic Description: The instrument gathers countrywide information on the nutritional status of the population, based on anthropometric (EPM), VAD, IDA and IDD indicators. [It covers family and individual information on women/caretakers and pre-school aged children: household sanitation; parental schooling; marital status; work status; occupation; mother/caretaker pregnancy status; number of previous pregnancies; breast-feeding practices; use and source of iron supplements and anti-helminthics; food items consumed in previous 24 hours; weight, height and hemoglobin; child's health status, recent morbidity, use and source of vitamin A and iron supplements and anti-helminthics, food items consumed in previous 24 hours, weight, height and hemoglobin. Other laboratory tests (e.g. acute phase proteins, micronutrient content of food samples) are optional. The current instrument does not include IDD information other than iodine content of table salt from households, as VAD and IDA surveys often focus on pre-school age children and women in reproductive age whereas the target group for IDD assessments (e.g. goiter, urinary iodine excretion) is usually the school-aged population.]
Country Applications: Honduras, Morocco, Zambia, and Mozambique.
Languages: Developed originally in Spanish. English and French modified versions used in Zambia and Morocco.
Technical Scope: To assess micronutrient status among women and children at national and regional (eventually district) levels as the basis for developing policy and program strategies for controlling micronutrient deficiencies and EPM.
Purpose: Baseline and future impact evaluations.
Type of Methods: Quantitative and semi-quantitative through household interviews of mothers/caretakers, as well as laboratory data.
Design: Cross-sectional design using a representative sample of the population of mothers and children within a specified age range, either national or from a defined geographic area. The instrument is used for collecting information to assess the existing situation regarding vitamin A deficiency (VAD) and iron deficiency anemia (IDA), as well as general energy-protein malnutrition (EPM) in children and their mothers or caretakers, and their relationships with epidemiological indicators.
Frequency of Administration: Five years from original survey.
Key Users of the Information
Primary Users: Governments, particularly Ministries of the Social Sector (Health, Education), as well as Ministries of Agriculture and Economy.
Secondary Users: Private sector, including NGOs/PVOs, food and pharmaceutical industries. USAID's missions, Cooperating Agencies (CAs), and other international donors (WHO, UNICEF).
Objectives and Scope of Tool
Objective: Gather information needed for assessing the magnitude and epidemiological factors related to VAD, IDA and, eventually, IDD and EPM, in representative samples of mothers and children, as well as for estimating current population coverage of micronutrient interventions (supplementation, food fortification).
Scope: The tool is used to collect and analyze nutrition/micronutrient problem identification information useful for policy development and program evaluation, as well as for decision-making regarding priorities to geographic areas and population groups, and improved intervention strategies.
Key Monitoring Needs and Evaluation Questions Tool Seeks to Address
The following questions are answered at the country level:
Key Indicators
Prevalence of VAD, IDA and EPM:
Food consumption:
Program coverage:
Research Design
The assessment survey is conducted in a statistically representative sample of eligible households, i.e. those with mothers and children within a specified age range. A Household Master Sampling Frame (MSF) is often available from which a nationally or regionally representative sample of households can be selected. The MSF may be divided in geographic strata (urban/rural, district, and sub-district). Usually the whole country is covered, except the isolated areas with low population density. Sample size is estimated on the basis of the expected prevalence of vitamin A deficiency. A sample of about 300 children per unit of analysis (region, district), with an estimated 20% non-response rate, provides valid prevalence estimations of around 20% with a 95% confidence interval and a 5% sampling error. A two-stage sampling procedure is recommended by which clusters and then households within clusters are selected, using a systematic selection method with probability proportional to the number of households. One or more eligible children per family/household may be covered, with sampling errors estimated on the basis of individual selection probabilities. The instrument does not include specific guidelines for sample design and selection, which may be found elsewhere.
Implementation
Survey implementation is logistically complex and requires careful planning and preparation. Usually a local Technical Committee (often set up by the MOH) reviews all plans and procedures. Conventional ethical regulations, including USAID's, are followed, even in the absence of a local Ethical Review Committee (e.g. written informed consent is obtained from mothers/caretakers). The questionnaire is pre-tested in both urban and rural households prior to final printing. Training and implementation manuals are prepared.
Interviewers training and standardization (questionnaire, anthropometric measures and blood collection, handling, storage and transportation) may take 2-3 weeks, and includes a pilot test (in households other than those selected) before formally initiating fieldwork. The target population is timely informed about the importance, content, scope and dates of the survey usually through radio and district/local health authorities. A data quality control system is set up by which a certain proportion of the interviews are observed by the supervisor, coding of questionnaires is carefully reviewed on a daily basis, and a certain proportion of blood samples are collected and analyzed in duplicate.
For a total sample of around 2,000 households, data collection may take from two to four months, depending on the local conditions (country extension, geographic dispersion of households, accessibility, season, type and status of roads, transportation facilities, and health districts and local support) and the number of interviewers. A field team comprised of one supervisor, two phlebotomists, and two interviewers may cover ten households per workday (five per interviewer/phlebotomist) in rural/disperse areas, and about twice that number in urban/dense areas. Adequate time should be allowed for field workers transportation and days off.
Hemoglobin is measured on site by inserting a drop of blood taken by finger-prick into a portable photometer with digital reading (Hemocue), and the result is directly recorded in the questionnaire. As specific blood spot tests are not available yet, venous blood specimens for laboratory tests other than hemoglobin (e.g. serum retinol) are collected, processed, preserved and transported to central laboratory facilities in the country or abroad; for faster and more reliable results, regionally accredited laboratory facilities may be used rather than building no existing local capacity.
The survey is often partially funded by the local government, sometimes with significant additional contributions from donors. Total cost may range from $ 100,000 to $ 200,000, excluding external technical assistance, contingent upon survey's scope and content, sample size, local salaries, duration of field work and transportation costs. A relatively high proportion goes to local salaries and transportation costs (travel, per diem); in some cases, laboratory equipment and supplies may be a significant cost item.
Analysis
Data entry and cleaning are best done as soon as possible after collection so that immediate verification of dubious/unclear information is possible. EPI-INFO database and cleaning procedures can be used. Additional time needed for completion of data cleaning and file preparation for analysis ranges from two to three months depending on the efficiency of the data quality control system in the field and of data entry procedures. Depending on the availability of local resources, the clean data files may be analyzed locally or abroad. Although data analysis should ideally be done locally with appropriate technical support, the need for rapid analysis and report preparation may require external assistance. At any rate, data interpretation and report preparation is best done locally.
In analyzing weight and height data, the NCHS/CDC/WHO reference values and cut-off point at <2 Z-scores for anthropometric indicators (weight/age, height/age, weight/height) are used, as well as the WHO recommended exclusion criteria for anthropometric values. Internationally accepted cut-off points for biochemical indicators are used for prevalence estimations. Hemoglobin values are adjusted for altitude (at the sea level) using the formula proposed by CDC. Acute phase proteins with conventional cut-off points, used as markers of current or recent infection in children, would allow estimating adjusted VAD and IDA prevalence in the absence of infection. Multiple regression analyses are used for identifying variables contributing to observed prevalence rates.
Reporting
Preparation of the survey report should be a responsibility of the local implementing institutions. Exceptionally, a preliminary survey report may be drafted abroad and used by the local institutions as a basis for preparing the official survey report, often complemented with general and specific recommendations. A prototype table of contents for the survey report has not been prepared as the content may depend on the survey scope. The usual content includes: background; survey objectives and methods (sampling frame, selection procedures, and methods for data collection, quality control, entry, cleaning and analysis); general demographic, social/economic and sanitary conditions of the households surveyed; reported child's morbidity; child's anthropometric indicators, vitamin A status and anemia; mother's anthropometry indicators and anemia; vitamin A content of sugar, and iodine content of salt. Host countries often request technical and/or financial assistance for report preparation and publication.
Dissemination of Results
The final official survey report, in the local language, is printed, released and disseminated by the local authorities, usually by the MOH. Report distribution is often complemented by formal presentations and discussions of the survey findings in technical planning workshops and seminars attended by key representatives from the public and private sectors, where problems are prioritized and general and specific recommendations made. Ultimately, the country authorities make policy and program decisions. Specific papers with summaries of the most relevant survey findings may be submitted for publication to scientific journals. However, by far the most important dissemination approach is through discussion meetings with decision-makers. According to USAID's regulations, once the final report is released survey files become public domain.
Manuals and Guidelines
Training manuals and field operation/data collection guidelines, including standard techniques for anthropometry measurements, and blood taking, handling, storage and transportation, as well as for laboratory assays, are prepared in the local language and field-tested before formal survey implementation. General technical guidelines and procedures for hemoglobin assessment are available from the Hemocue manufacturer (in English, Spanish and French) and from USAID projects (OMNI, MotherCare). Detailed manuals on quality control for blood spot hemoglobin assessment with Hemocue are available from PATH.
References
Encuesta Nacional de Micronutrientes. Honduras, 1996. Ministerio de Salud de Honduras, Fundacion Internacional de Ojos (FIO), USAID/OMNI. Tegucigalpa, Honduras, 1997.
Nestel P, Melara A, Rosado J, Mora JO. Undernutrition among Honduran children 12-71 months of age. (Submitted for publication, 1998)
Nestel P, Melara, Rosado J, Mora JO. Vitamin A deficiency and anemia among children 12-71 months of age in Honduras. (Submitted for publication, 1998)
Nestel P, Melara, Rosado J, Mora JO. Nutrition of Honduran mothers/caretakers. (Submitted for publication, 1998).
Other
At present, there is no other prototype instrument available for national VAD, IDA and IDD or EPM prevalence assessments in developing countries. An updated field manual for survey planning and implementation is badly needed; a Manual for Nutrition Surveys prepared in the early 1960s by ICNND is still being used in some countries.