Cover Image
close this bookMethods for the Evaluation of the Impact of Food and Nutrition Programmes (UNU, 1984, 287 pages)
close this folder6. Measuring impact by assessing dietary intake and food consumption
View the document(introductory text...)
View the documentReasons for assessing household food practices
View the documentDeciding on a method
View the documentModels for household dietary assessment
View the documentAdditional data
View the documentSteps and procedures in surveying
View the documentAssessing dietary change
View the documentSummary
View the documentAnnex A. Development of a score card for the rapid assessment of calorie and nutrient intakes
View the documentAnnex B. Household quantitative intake record
View the documentReferences

Models for household dietary assessment

Models commonly used include qualitative and quantitative intakes, written records and oral histories, complete household intakes, and intakes of certain foods only. These cover periods of time ranging from one day to an entire season or longer. Table 6.1 summarizes some of the models. A discussion of these methods follows.


Initial Informal Assessment

Before attempting a formal dietary survey, some experienced team member or members should live in the community for at least a few days to observe the total food situation. They should visit food markets and stores; converse with people; observe gardening and other food production practices; note methods of food preparation; and eat local dishes-if feasible, in people's homes. Observers can gain an impression of the degree of interhousehold and day-to-day intrahousehold variation. They can gather information about such concerns as buying practices, market days, pay days, and animal slaughtering days. They can determine standard portion size for certain foods, e.g., loaves of bread. Much information can be obtained that can subsequently be used for evaluation purposes, and so may even eliminate the need for more formal surveys. Even when formal surveys are indicated, anecdotal records can greatly enhance the value of statistical data. Often the community itself and also outside authorities can be persuaded of the need for change on the basis of appropriate anecdotes.


Determining Household Intake of Calories and Many or all Major Nutrients

While data in food terms only may be useful for food production and other programmes, some nutrition programmes require conversion of quantitative food intake into the common denominators of calories and nutrients. If time, money, personnel, and circumstances permit, one may want to do as complete an assessment as possible of the total dietary structure in order to help uncover unsuspected deficiencies or irregularities in the diet. Four methods that have been used are:

  1. Weighing and measuring of all food for as extended a period of time as feasible or necessary (usually a week) to obtain a fair measure of the "usual" intake. This method, designated here as the "detailed method," has been described in an early FAO publication (1).
  2. The detailed dietary history (2).
  3. A more rapid modified dietary history method (see Annex B).
  4. The food inventory method, used by the United States Department of Agriculture in its early household food consumption studies (3).

The detailed method of weighing or measuring food for a week involves a daily visit to each home in the sample to weigh or measure all food to be consumed on that day. Corrections for later additions and for left-overs must then be made in a subsequent visit. This method obviously requires much interviewer time and has the further disadvantage that subjects may modify their diets during the recordkeeping period. It has the obvious advantage of greater validity. In a functionally literate population, the homemaker may keep a record of food consumed. Frequent checking of such records by a trained investigator is essential.

The detailed quantitative dietary history, covering periods of time ranging from a week to a lifetime, requires skill and patience on the part of both interviewer and client. It is commonly used for research purposes only.

TABLE 6.1 Methods of Household Dietary Assessment and Some of Their Advantages and Shortcomings

Method Techniques Advantages / Use Shortcomings
1. Initial informal assessment. Survey personnel live in community to observe food pratices and gather preliminary data. Sets stage for later survey facilitates planning. Subjective impressions may be inaccurate or incomplete.
2. Detailed dietary history (long term). Inteviewer inquires in detail about food consumed over specified time period. Quantitative assessments made. More commonly used for individual than household intakes. Accurate data: Only quantitative method available to cover long time periods Usually used for research purposes. Requires trained. extremely skillful interviewers and intelligent. cooperative subjects. Time consuming and costly.
3. Weighed or measured food intake ("detailed method"). All household food for specified period (usually a week) weighed or measured and recorded by interviewer or householder. Accurate data. if well gathered can help explain biochemical and clinical date. Readily converted into nutrients. Intake may be influenced by the record-keeping process. Time consuming and expensive.
4. Modified dietary history method. Inteviewer assesses quantitatively the household food intake for past week or days. Reasonably accurate data. Quicker and less expensive than weighed intake and detailed history. Can be converted into nutrients. Method of choice if complete data are needed. Period covered may not be typical of long-term diet. Fairly time-consuming and costly. Requires fair degree of knowledge and skill on part of interviewer.
5. Food inventory method. All food in household weighed at beginning and end of time period. All food entering or leaving household recorded. Accurate data. Method not much used any longer because of time and effort involved. Costly and time-consuming. Method not relevant in areas where there are no food stores.
6. "Score card" method. Interviewer assesses quantitative intake of those foods high in specified nutrients Calculation of nutrient intake is only. "Points" assigned for specified quantities. Shorter interview time because not all foods are assessed based on a previous survey or quick and easy. Data can be processed in the field. Relatively inexpensive. Useful if 4 is not possible. Information contained is limited to certain nutrients. Method is considerable knowledge of community food practices.
7. 24-hour diet history. Same as 2, but covers only 24-hr. period. Much faster than 2. Requires less skill on part of interviewer and client. Best used for assessing community dietary status rather than individual household because of interhousehold variation. Intake may not be "typical" of "usual" intake, especially in regard to quantities.
8. Food frequency lists. Householders report frequency of consumption of foods on a list. Quick and easy. Gives picture of food patterns. Non-quantitative. Cannot be used to assess nutrient intakes.
9. "Scalograin" method. Householders report frequency of consumption of foods that have "scale" value. Quick and easy. Permits classifying population into consumption strata. Non-quantitative. Does not assess nutrient intakes.
10. Single food intakes. Householders report intake (usually quantitative) of one food only. Quick and easy. May indicate use of recently introduced food. or of a particularly desirable or undesirable food. Limited to the food in question. Does not assess nutrient intake.
11. "Indicator" foods. Householders report use of foods indicative of total food pattern. Quick and easy. For designating food patterns. Useful only to indicate possible problems.


The more rapid modified dietary history method is based on a single interview during which the interrogator tries to obtain the "usual" daily food intake of the family, and specific intakes for as many days of the time period under consideration as the homemaker can recall. Quantities are then corroborated by inquiries about food purchases and omissions are corrected through use of a "reminder list," listing all foods common to the area. From these interrogations the interviewer then constructs an average daily intake for the time period under consideration. Such an interview may require several hours, depending on the complexity of the diet, the interviewer's skill, ability of the homemaker to recall, etc. It is, of course, subject to the inaccuracies of recall. Skillful interviewing can mitigate this problem. Compared to record-keeping, it has the advantage of not enabling subjects to modify their diet. Validity of information obtained is influenced greatly by the degree of rapport established, clients' understanding of the purpose of the survey, and so forth. It has been successfully used to obtain reasonably complete information when it was not practical to carry out either a detailed assessment of dietary intake or a detailed dietary history.

The food inventory method involves taking an inventory of all food in the household at the beginning and end of the survey period as well as records of food entering the household during the period and of food wasted or fed to animals. While dependable, (especially when literate clients may assist in record-keeping), this method, like the FAO method, requires much interviewing and processing time. Furthermore, record keeping, whether done by clients themselves or interviewers, has the disadvantage of permitting clients to change their food intake during the record-keeping period. It has the advantage of relative validity.


Determining Intake of Calories or a Limited Number of Nutrients Only

Sometimes, as in situations of obvious food scarcity, only calories need to be assessed. For this purpose, measurement of just the staple food is usually adequate. In other situations one may want to assess only a few specified nutrients. Not all programmes need to assess all nutrients. If, for example, it is obvious that all households are consuming rich sources of vitamin C, and biochemical and clinical data bear out this observation, one may not want to assess fruit and vegetable intake. If, on the other hand, there is evidence of protein-calorie malnutrition, one may focus on main calorie and protein sources only. In such instances the "Score Card Method" (see Annex A) may be practical. (In our terms, "score card" implies a shortened nutrient calculation method.) The method does require preliminary study to determine the chief sources of calories and protein in the "usual" dietary. Such a study may be a small one carried out by a nutritionist in advance of the proposed major survey. When this method is used, only foods that are major sources of the nutrients in question need to be assessed, although amounts must be determined because the food data are to be converted into quantative nutrient terms.

Compared to the three more detailed methods previously outlined, the score card method saves considerable interview time and also nutrient calculation time. When tested in Guatemala, it showed approximately 70 per cent agreement with the modified dietary history method (Annex B) for calories and protein, 60 per cent for iron, and 82 per cent for vitamin A.


24-hour History

By this method, the interviewer determines the previous 24-hour food intake of the household, usually in quantative terms. In a literate community, homemakers may keep a record of such an intake.

Because of the short time period covered, this method is more economical than the "detailed method" and the modified dietary history method. One day may or may not represent a "typical" intake for the individual household. Twenty-four-hour intakes of a large sample of households may, however, represent a typical daily intake for the community as a whole.


Food Frequency Lists

By this method, a client is asked simply how frequently the household consumes specified foods from a list of those available in the community. For some purposes, such as determining the popularity of specified foods, this information is sufficient. Usually no attempt is made at quantification.

The food frequency method has the obvious advantage of ease, rapidity, and consequent low cost. A main disadvantage is that non-quantitative data cannot be converted into nutrient terms. A few assumptions about data showing only frequency of food consumption seem warranted, however:

  1. A monotonous diet consisting of one or two foods only is at risk of being deficient in calories and nutrients.
  2. A diet extremely low in fat is at risk of being calorically deficient.
  3. If consumption of fruits and vegetables is seasonal, intakes of vitamins A and C are likely to be low at certain times of the year.


Scalogram Method

The scalogram method is a variation of the food frequency list method, but only those foods known to have "scale" value are included (4). "Scale" foods are those whose consumption is closely linked to socio-cultural factors and therefore define a consumption stratum. By means of a formula designed by Guttman, foods can be defined as having scale value. The method is relatively rapid and easy. It permits determining the proportion of a population falling into specified consumption strata and therefore their relative risk of malnutrition. Population consumption pyramids of communities can be compared with each other to identify those at greatest risk.


Assessment of Household Consumption of a Single Food Product

One may wish to gain insight into the consumption of (a) a food recently introduced into the culture, e.g., a corn-soya product (b) a specific food or food group already in the dietary or (c) an "indicator food".

Concerning the case of consumption of new food, amounts used may be obtained from dealers or distributors. Furthermore, if the number of households consuming the product and the range of consumption is to be assessed, a sample of households can be interviewed directly. The interview should include a few tactful questions regarding the product to ascertain whether or not it is indeed being used for human consumption and by whom. Caution: one dare not assume from such consumption data alone that the product constituted an addition to the diet; it may well have served as a substitute for other items previously consumed. Careful interviewing may elicit this information, as may comparison with baseline survey data if available.

As for the case of specific food or food group already in the dietary, one may want to determine the general acceptance of certain foods or food groups for nutritional evaluation or programme reasons. If, for example, certain vegetables are found to be popular, even though underused, they might be incorporated into a gardening programme; or similarly, small animal production may be indicated if meat is popular. Total absence of fruits and vegetables may help to explain clinical and biochemical findings indicative of the lack of ascorbic acid. A short interview could obtain such data, either qualitatively alone. or quantitatively if needed.

The third category mentioned above, indicator foods, are those single foods or food groups which may be indicative of a total food pattern. For example, in Mexico the use of bread or rolls rather than tortillas is likely to indicate adoption of non-indigenous eating practices. Consumption of indicator foods may be a useful way of assessing change, although it is not an assured way of determining nutritional adequacy.