D. Food and Nutrition

The information in this section is designed to help Assessment Team members understand and evaluate nutritional data from surveys they may encounter in the field. It is not intended to teach them how to conduct these activities. Context-specific information is also needed in the survey to explain the rate of malnutrition. For example, if the rate of malnutrition is very high, the team should ask if a disease outbreak occurred before the date of the survey. A diarrheal disease, acute respiratory infection, or measles outbreak that occurred in the 2 weeks preceding the survey may help explain the increase in malnutrition.

Followup surveillance of the population as a whole should be done by gathering information about the nutritional status of the children, using the WFH or weight-for-length (WFL) comparison method. WFH is a lagging indicator; by the time it shows a significant change, the situation has already deteriorated. Children are the first to show signs of malnourishment during a food shortage. Their nutritional status is an indicator of the amount and degree of malnutrition in the population as a whole. Using the WFH or WFL comparison method on a random sample of children will help assess the nutritional status of the population as a whole.

When using the WFH or WFL method in a surveillance program, a random sample of children from 6 months to 59 months of age (under 5 years of age) or less than 110 cm tall are weighed and measured regularly. Children less than 85 cm tall are measured supine (WFL) and children above 85 cm tall are measured standing up (WFH). See the description of the weight-for-height (weight-for-length) method in the next section for additional information on assessing malnutrition using this method.

a. Sampling Methods

In a given population, 18 to 20 percent will be under 5 years of age. For a small displaced population (2,000 to 3,000), all the children, about 400 to 500, should be measured. For larger populations, two methods of surveillance are most applicable: systematic random sampling and cluster sampling.

(1) Systematic Random Sampling

Systematic random sampling, in which an interval of every n^th shelter or household is sampled and all children under 5 years of age in that household are measured, is recommended where populations are living in an organized or structured setting as a camp. To undertake a systematic random sampling, one needs to know the total number of households, the average number of children under 5 years of age in a household, and the total population. The recommended sample size for a systematic random sampling is 500 children. This sample size will ensure 95 percent probability that the sample is representative.

To calculate the n^th interval, first calculate the number of households to be visited to measure 500 children. Use the following equation: 500 / (A x P), where A = the average household size and P = the proportion of children under 5 years of age in the total population.

For example, if the average household size is six persons and the proportion of children under 5 years of age (110 cm in height) is 17 percent, then 500 / (6 x 0.17) = 490 households to be visited to measure 500 children.

If 10,000 households reside in the population, the n^th sampling interval would equal 10,000 / 490 households to be visited = 20. Therefore every 20th household should be visited until 500 children under 5 years (110 cm) are measured.

(2) Cluster Sampling

Cluster sampling, in which random "clusters" of households are measured, is used in populations that are not in even or structured settings but are spread unevenly over a large geographic area. The cluster sample method is most often used because, in the initial stages of emergencies, people are rarely living in a structured pattern. The sample size needed to obtain 95-percent probability is 900 children. For reliable results, examine at least 30 clusters and not less than 900 children or "30 clusters of 30."

In a rapid assessment, the area of interest should be divided on a map into sections of equal size. Each section should have at least 300 inhabitants. The total number of clusters, drawn from a list of all sections and their population estimates, is divided by 30 to obtain the cluster interval k. Starting at a random selected interval, every k^th cluster is selected. For example, suppose that there are a total of 210 clusters. This total number of clusters is divided by 30 to obtain the cluster interval (210 / 30 = 7). Starting from a randomly selected cluster, every 7^th cluster on the list is selected until 30 survey clusters are chosen.

During the survey, the team starts at the center of the cluster and chooses a direction (by spinning a bottle or pen). The survey is then started at the next nearest household or shelter in that direction. All eligible children in the household are measured. The team moves to successive houses until 30 children have been examined.

Pay particular attention to the geographic area or food economy groups in which people live, especially in drought emergencies. For example, if a large geographic area is inhabited by pastoralist, agropastoralist, and agriculturalist groups, a separate survey should be conducted for each group, as the effects can be quite different among these groups.

Initially, such surveys should be conducted every 2 months. When conditions have stabilized, a survey once every 3 to 6 months is sufficient. Any change or trend in nutritional status can thus be detected and adjustments made in the relevant feeding programs.

If the initial assessment indicates a need for supplementary or therapeutic feeding, individuals with these requirements should be identified and registered for appropriate programs. Their individual progress should be monitored through more frequent weighing at feeding centers.

b. Malnutrition Measurement Methods

Malnutrition can be recognized by certain clinical signs (e.g., marasmus, kwashiorkor, and marasmic-kwashiorkor) and body measurements. Body measurements are required for objective assessment of nutritional status and comparison with regular surveillance data.

Acute malnutrition is measured by the WFH method, while chronic malnutrition or stunting is measured by height for age. In many areas, the information collected at health facilities is the rate of stunting rather than acute malnutrition. OFDA’s responsibility is to decrease acute malnutrition to save lives.

(1) Weight-for-Height (Weight-for-Length) Method

The weight-for-height (weight-for-length) measurement method, which is expressed either as a percentage of a reference median or as a Z-Score, is preferred for nutritional surveillance and for measuring individual progress in emergencies. If a percentage is used, it indicates the weight of the child expressed as a percentage of that of a well-nourished child of the same height as given in international reference tables. If a Z-Score is used, the "Z" represents the median for children and the Z-Score represents the number of standard deviations above or below the median (because the population is normally distributed, the median equals the population mean). Children with less than 80 percent weight-for-height or with a Z-Score of less than - 2 are classified as malnourished; those with less than 70 percent weight-for-height or with a Z-Score of less than - 3 are considered severely malnourished. Without special feeding programs, severely malnourished children will die.

Note that the weight-for-height method should not be used for children who have edema; they should automatically be classified as severely malnourished.

Abbreviated reference tables with weight-for-height and weight-for-length comparisons and Z-scores are located in tables III-4, III-5, III-6, and III-7.

Table III-4. Weight-for-Length Expressed as a Percentage of Median Weight (length assessed supine up to 85.0 cm, sexes combined)

Table III-5. Weight-for-Height Expressed as a Percentage of Median Weight (height assessed by standing from 85.0 cm, sexes combined)

Table III-6. Weight-for-Length and Associated Z-Scores (length assessed supine up to 85.0 cm, sexes combined)

Table III-7. Weight-for-Height and Associated Z-Scores (height assessed by standing from 85.0 cm, sexes combined)

(2) MUAC Method

Another method used when a rapid screening of young children is necessary is the MUAC measurement. This measurement is less precise than the WFH method but can be done more quickly. MUAC is an indicator of the risk of death rather than malnutrition. It can be used in insecure areas where access to the vulnerable population is very limited. MUAC measures the part of the arm where the circumference does not normally change significantly between the ages of 1 and 5, but wastes rapidly with malnutrition. The technique is not suitable for monitoring the progress of individual children. Professional help should be used for the MUAC method.

If the child has no signs of edema, the MUAC of the child should be measured, using a measuring tape specifically designed for MUAC measurements. The tape should be wrapped closely, but not tightly, around the arm, midway between the elbow and the point of the shoulder. The arm should be hanging loosely, with the tape measure circled around the arm. The tip of the tape measure should be inserted back-to-front through the narrow slit at the white end of the tape. The arm circumference should be read to the nearest 0.1 cm between the vertical arrows at the center of the large opening. The arm circumference of normal children between 1 and 5 years of age changes very little. Therefore, children of these ages can be included in a nutrition survey using the same standards.

The arm circumference tapes have colored bands representing the different nutritional states below.

If custommade measuring tapes are not available, a thin strip of plastic (about 30 cm in length) should be used with marks clearly indicated at the zero point, 12.0 cm, 12.5 cm, and 13.5 cm.

Each arm circumference has an approximate equivalent to a WFH percentage.

• 13.5 cm or greater: approximately equivalent to more than 85 percent WFH.
• 12.5 to 13.4 cm: approximately equivalent to 80 to 85 percent WFH.
• Less than 12.5 cm: approximately equivalent to less than 80 percent WFH.

The amount and degree of malnutrition can be calculated as percentages of the sample. The percentage of children with edema (kwashiorkor) should also be reported.

The child should be checked for edema, the swelling associated with kwashiorkor, before being measured using either the WFH or MUAC method. Press a finger against the front of the child’s foot for about 3 seconds. A dent (pitting) indicates that the child has edema and therefore should not be measured. It should be recorded that the child has edema and is severely malnourished.

(3) Interpretation of Malnutrition Rates and Corresponding Actions

^_a Malnutrition rate is defined as the percentage of the child population (6 months to 5 years) who are below either the reference median WFH - 2 SD or 80% of reference WFH.

^b Aggravating factors:

• General food ration below the mean energy requirement
• Crude mortality rate more than 1 per 10,000 per day
• Epidemic of measles or whooping cough (pertussis)
• High incidence of respiratory or diarrheal diseases

Prev: 4. Nutritional Diseases
Next: 6. General Feeding Programs
TOC: D. Food and Nutrition