Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

A simplified tool for assessing vitamin-A deficiency risk at the community level

Miriam M. de Chávez Homero Martínez and Adolfo Chávez

Vitamin-A deficiency has recently been shown to have an impact on morbidity and mortality [1, 2, 3]. The extent of eye lesions has been the method commonly used to assess the severity of vitamin-A deficiency [4, 5]. While this method has a very good correlation with serum vitamin-A status and has been widely used in field studies, it requires specialized and well-trained personnel [6, 7]. Also, there is the need to examine large groups of at-risk individuals to find evidence of the deficiency. While the clinical examination provides very good information in warm, tropical climates, where the eye lesions detected can generally be assumed to be due to vitamin-A deficiency, in other types of environment, such as a medium or high altitude, or dry and dusty places, corneal reactions to the environment may be difficult to separate from early-stage vitamin-A-deficiency lesions.

Biochemical assessment of vitamin A is a more accurate way of documenting the vitamin-A status of the individual. However, its use at the community level is limited by the difficulties of drawing blood samples from large groups of individuals due to economic, logistical, and ethical considerations. As an alternative, a dietary task group of the IVACG has proposed the use of a dietary questionnaire applied at the home level by auxiliary health personnel with minimum training as a simplified method to evaluate vitamin-A-deficiency risk [8].

The dietary questionnaire

We have designed a simple qualitative questionnaire to evaluate the risk of finding vitamin-A deficiency in the community, intended to be administered at the home level during a 10-minute visit. The questionnaire is to be answered by the mother and is focused on pre-school-age children as the target population.

The information collected refers to the child's habitual diet, asking the mother about specific food items containing vitamin A consumed by the child on a weekly basis. The food items are placed in one of three groups according to their retinol content, based on information from Mexican food-composition tables [9]. Group 1 consists of foods with a retinol content Ł 19 mEq per 100 g; group 2, foods with 20-124 mEq of retinol per 100 g; and group 3, foods with ł 125 mEq of retinol per 100 g. A fourth term, "positive feeding habits," was established to include food habits that enhance vitamin-A absorption and/or utilization by the child. An arbitrary scoring system was devised in order to identify those with low, medium, and high consumption of vitamin A or its equivalent. The score is based on giving one point for each food consumed from group 1, three points for each food from group 2, and five points for each food from group 3 or for each positive feeding habit in the child's diet.

Study population and methods

A study to evaluate the dietary questionnaire was conducted in two parts, applying it to two different populations on which the Instituto Nacional de la Nutrición (INN) already had information regarding vitamin-A intake.

The first part was devised as a pilot testing of the questionnaire to determine whether the proposed scoring could discriminate between families with different levels of vitamin-A intake. For this purpose we used data from a study conducted by the INN six years earlier on the Yucatan Peninsula in southeastern Mexico, where vitamin-A deficiency is endemic. That study had shown that the dietary intake of vitamin A was very low [10]. For this stage we defined as the study population 310 families that had been studied in the previous project, and selected a random sample of 50 habitual-diet interviews from those involving families with children under five years of age. A nutritionist at the INN took the information required to fill in our vitamin-A-deficiency risk questionnaire from those interviews.

The second part of the study consisted of a field trial of the questionnaire in Calderas, a community in the Solis Valley in the central highlands of Mexico, where the INN has a rural research station. The dietary intake of the people in the Solis Valley had been previously documented to be low in vitamin-A content throughout the year, but clinical vitamin-A deficiency is not prevalent in this area [11].

On the basis of the census of the population of Calderas, a random sample of 50 households with children under five years old was drawn. These homes were visited by a field nutritionist to invite the mothers to participate in the study. All the mothers were invited to bring their child(ren) to the community health centre to have a blood sample taken. Besides the 50 children selected at random, 11 of their elder siblings were brought voluntarily by their mothers to the centre the day this activity took place and were also included in the study. Once the blood sample was drawn, the dietary questionnaire was administered by the field nutritionist or an auxiliary health worker who had been trained in the use of the questionnaire by the nutritionist.

The purpose of the blood sample was to determine the child's serum-retinol values in order to validate the score obtained from the questionnaire. Mothers willing to have their child participate in the study were asked to sign an institutional informed-consent form. The blood samples were drawn from a peripheral vein by members of a term of trained physicians assisted by auxiliary nurses. As a compensation for the child's participation in the study, we also determined (1) anaemia according to haemoglobin and haematocrit values in the blood samples, and (2) intestinal parasites by taking a stool sample. Positive cases for any of the conditions tested were offered free treatment, with a physician delivering the prescription and the drug(s) to the home of the patient.



The mean daily retinol intake of the 310 families in Yucatan (the study population), according to the 1982 survey, is shown in table 1. Ninety-six per cent of these families were taking less than their recommended daily allowance of vitamin A.

TABLE 1. Distribution of families in Yucatan according to their daily retinol intake (N = 310)

Intake per person (µg)

Families (%)









TABLE 2. Correspondence between percentages of children in the Yucatan sample according to their serum-retinol status and questionnaire scores (N = 50)

Serum-retinol status

Questionnaire scores

Retinol (µg)

% of children

Score (points)

% of children





1 1-20









The 50 children who formed the sample population were divided into three groups according to their serum-retinol status, and the percentage of children in each group was correlated with the corresponding percentage of scores obtained from the questionnaires that were applied to this sample. The correspondence between them is shown in table 2.

The percentile distribution of the total scores obtained from the 50 dietary interviews in Yucatan is shown in figure 1. The scores ranged from 7 to 29 points, with a mean of 19 + 5 SD. The percentage contributed to the total score by each food group is shown in table 3.

FIG. 1 Percentile distribution of the questionnaire scores for children in Yucatan obtained from dietary interviews (N = 50)

TABLE 3. Relative contribution (percentage) of the different food groups to the total questionnaire scores in the sample populations


Group 1

Group 2

Group 3

Positive feeding habits












Solis Valley

A total of 61 questionnaires were administered in the community of Calderas to children under eight years old. The nutritionist conducted 46 interviews, and the auxiliary worker conducted another 15. The percentile distribution of the total scores of the questionnaires administered by the nutritionist is shown in figure 2. The scores for the whole population ranged from 27 to 117, with a mean of 69.6 + 21.8 SD. The relative contribution of each food group to the total score is shown in table 3.

From a comparison of the scores obtained by the nutritionist and those obtained by the auxiliary worker in Calderas, it is evident that the nutritionist consistently obtained higher scores than the auxiliary worker, as may be seen in figure 3.

FIG. 2. Percentile distribution of the questionnaire scores obtained by the nutritionist in Calderas (N = 46)

FIG. 3. Comparison of questionnaire scores obtained by the nutritionist (N = 46) and by the auxiliary worker (N = 15) in Calderas



The results of this study support the use of the proposed questionnaire as a good screening tool to identify populations with low vitamin-A dietary intake. According to our expectation, based on previous dietary surveys in the two areas chosen to test the questionnaire, scores obtained in the Solis Valley, a non-endemic area for xerophthalmia/vitamin-A deficiency, were notably higher than those found in Yucatan, where clinical vitamin-A deficiency is highly prevalent.

Although there is a well documented cyclic pattern of ingestion of foods containing vitamin A during the year, and a one-time application of the questionnaire will only give a cross-sectional view of the population's dietary intake, we feel that in order for a community to be considered at risk of clinical vitamin-A deficiency the scores obtained by the dietary interview would have to be excessively and consistently low, as is the case in the Yucatan peninsula. Where there is a wide range in the dietary intake scores, as in the Solis Valley, the risk of vitamin-A deficiency is certainly lower.

An interesting feature of the questionnaire is the breakdown into the four food groups previously described for their contribution to the total score of vitamin-A intake. This breakdown allowed us to evaluate the relative contribution of each of the food groups to the total score. In the case of Yucatan, we saw that vitamin-A intake is mainly due to the ingestion of food items from group 1 and that the dietary habits that promote vitamin-A absorption also contributed significantly to the total score. In the Solis Valley the highest contribution to the total score was from food items in group 3, which is composed mainly of animal sources of vitamin A.

These findings have several implications for nutrition-education programmes. For example, in the former population it is clear that most of the vitamin A consumed comes from vegetables (although their ingestion is still insufficient); so more positive stress should be placed on promoting consumption from the animal food group. At the same time, positive feeding habits that can enhance vitamin-A absorption need reinforcement. On the other hand, in the Solis Valley it seems to be clear that food items from animal sources are already being consumed, and that feeding habits that enhance vitamin-A absorption are also present, while more emphasis needs to be placed on promoting the consumption of vegetable sources of vitamin A. Of course, nutrition-education programmes should focus on those food items that are available in the community and that may be more culturally accepted.

Another objective of this study was to evaluate whether the questionnaire could be successfully applied by an auxiliary field worker with minimum training. The scores obtained by the auxiliary worker were consistently lower than those obtained by the nutritionist. We think that this difference was mainly due to the fact that nutritionists are trained to be more insistent in seeking information on the individual's diet, while the field worker may write down only what the interviewee tells her on a first question, without insisting more about dietary habits. Therefore, if the questionnaire is to be successfully applied by auxiliary workers, special care should be devoted to this issue during their training. Also, supervision during the early stages of the field work will play an important role in ensuring that the interviews are applied as expected.



This study was made possible by the interest and support of the IVACG. Special thanks should be given to the field staff who earned out the interviews and laboratory examinations in the Solis Valley.



1. Sommer A, Katz J, Tarwotjo I. Increased risk of respiratory disease and diarrhea in children with preexisting mild vitamin A deficiency. Am J Clin Nutr 1984; 40:1090.

2. Sommer A, Tarwotjo I, Kjunaedi E et al. Impact of vitamin A supplementation on childhood mortality. Lancet 1986;ii:1169.

3. Martinez H, Shekar M, Latham M. Vitamin A supplementation and child mortality. Lancet 1986;ii:451.

4. Sommer A. Field guide to the detection and control of xerophthalmia. 2nd ed. Geneva: WHO,1982.

5. Sommer A, Hussaini G, Muhilal H et al. History of night blindness: a simple tool for xerophthalmia screening. Am J Clin Nutr 1980;33:887.

6. Solon FS, Popkin BM, Fernandez TL, Latham MC. Vitamin A deficiency in the Philippines: a study of xerophthalmia in Cebu. Am J Clin Nutr 1978;33:360.

7. Blackfan KD, Wolback SB. Vitamin A deficiency in infants: a clinical and pathological study. J Pediatr 1933;3:679.

8. Proceedings of the IVACG meeting in Brasilia, Brazil, 1986.

9. Hernández M, Chávez A, Bourges H. Valor nutritivo de los alimentos mexicanos. 2nd ed. Publication L-12. Mexico City: Nutrition Division, Instituto Nacional de la Nutrición, 1981.

10. Mata A, Sandoval J, Geiger S et al. Diagnóstico sobre la dificiencia de nutrimentos en Yucatán y bases pare el programa de enriquecimiento del azúcar. Publication L-51. Mexico City: Nutrition Division, Instituto Nacional de la Nutrición, 1981.

11. Rodriguez G. Variaciones en el consumo de nutrimentos y de energía en un ciclo estacional en una comunidad rural. Thesis for the degree of Nutritionist and Dietitian. Mexico City: Secretaria de Educación Pública, 1987.

Contents - Previous - Next