Contents - Previous - Next

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

Mothers' knowledge, understanding, and use of the bubble chart in a rural area of central Mexico

Homero Martínez, Miriam Muñoz de Chávez, Noé Guarneros, Alicia Ríos, and Adolfo Chávez



The bubble chart is a growth chart that has a vertical layout rather than the more common horizontal one, and bubbles instead of cries-crossing lines. The present project aimed to evaluate knowledge, understanding, and use of the bubble chart by 85 mothers living in a rural area of central Mexico. The evaluation was done with a pre-intervention post-intervention design, and was based on Bloom's taxonomy of the learning process, which defines a sequence that goes from knowledge to understanding (comprising translation, interpretation, and extrapolation), and finally to application. All areas evaluated showed an increase between pre-intervention and post-intervention, with a statistically significant (p <.05) increase in knowledge, interpretation, and application; the mothers ranked highest in knowledge. The children's nutritional status increased significantly (p <.0001) between pre-intervention and post-intervention. Except for one child, they had at least one period of weight loss during the study period. However, 81% of them gained weight most of the time (i.e., at least two-thirds of records showed weight increments), 7% maintained their weight most of the time (at least two-thirds of weight records showed neither gain nor loss), and 12% lost weight most of the time (at least two-thirds of records showed weight loss). Mothers of children who mostly gained weight had higher knowledge, understanding, and application scores (p <.01). Similarly, these mothers were most likely to perform all the activities promoted as part of the growth programme (p <.001).


Child malnutrition is common in most areas of the developing world. Several classifications of child malnutrition have been developed based on different anthropometric indexes, mostly weight and height, related to each other or to age [1, 2]. Some authors have criticized them, however, on the basis that they do little to help prevent or cure malnutrition in children [3]. Monitoring growth by periodically (usually monthly) recording a child's weight on a chart that shows growth increments is useful in helping to prevent malnutrition by increasing mothers' awareness of children's weight gain and loss [4, 5]. By pointing out expected increases in weight or insufficient monthly gains, health workers inform mothers how deficient weight gain leads to malnutrition, infection, and eventually death [6]. However, to involve mothers effectively in the required interventions to prevent or treat malnutrition, they must learn to interpret the growth chart correctly.

To this end, different designs of weight charts have emerged since the concept of growth monitoring was first established and endorsed by UNICEF. One of the recent advances is the bubble chart [7, 8]. Basic modifications to the original chart consist of using bubbles instead of vertical and horizontal crossing lines to make it easier for health workers and mothers to locate the child's weight according to age, and using a vertical layout rather than a horizontal one (see FIG. 1. The bubble chart). A direct consequence is that the direction of the growth vector (i.e., the direction that the last weight takes in relation to the previous two weights) is magnified. Therefore, if the child is growing well (i.e., puts on weight), the growth vector will be positive (line pointing upward); if the child does not gain weight, it will be neutral (horizontal); and if the child loses weight, it will be negative. The direction of the growth vector helps the health auxiliary and the mother decide whether they should take specific actions in relation to the child's growth.

The purpose of this project was to evaluate the knowledge, use, and understanding of the bubble chart by mothers and health auxiliaries in a rural area of central Mexico.


Study area

The project was undertaken in five rural communities located in the valley of Solis in the central highland plateau 170 km northwest of Mexico City. This area shares several characteristics with other rural communities typical of central Mexico, such as a predominantly mestizo origin, Spanish language, Catholic religion, agriculture as the main activity, and maize and beans as the staple diet [9].


Project activities

The intervention was carried out by field auxiliaries. One in each of the five study communities. These were women from the community who were hired as full-time workers for the project. They participated in a structured training programme conducted by a multidisciplinary team (psychologist, anthropologist, physician, social worker). Team members also participated in field activities, providing close supervision of the auxiliaries' activities and reinforcement of positive attitudes, as well as conducting weekly meetings to review problems and strategies to gain mothers' participation in weighing their children.

Based on a demographic census of the communities, all mothers with children under two years of age at the beginning of the project were visited by the field auxiliaries and invited to participate in the programme on a voluntary basis. Mothers who accepted were visited at home once a month by the community field auxiliary. The number of houses that each auxiliary visited varied from 12 to 20. On their visits, they weighed the child, plotted the weight on the growth chart, and instructed the mother on how to interpret the chart. When growth faltering was detected, the mothers were advised on specific actions they could take to reverse it. These actions included reinforcing proactive feeding behaviours for small children, offering a more varied diet (within local constraints) to older children, and taking children to the doctor when an acute disease such as diarrhoea or upper respiratory tract infection was present. Children were classified as mostly "gainers" when at least two-thirds of their records showed weight gain, mostly "maintainers" when at least two-thirds of their records showed neither gain nor loss, and mostly "losers" when at least two-thirds of their records showed weight loss.


Theoretical framework for evaluating mothers' knowledge, understanding, and use of the growth chart

The evaluation of mothers' knowledge, understanding, and use of the growth chart was based on Bloom's taxonomy of the learning process [10]. According to Bloom, learning follows a sequential process from theory to practice. Knowledge is said to occur when an individual is able to reproduce a specific piece of information. Next comes understanding, which is divided into three levels: translation, in which an individual will be able to rephrase the information acquired; interpretation, meaning that a certain mental process takes place in such a way that a given piece of knowledge receives a treatment that makes it appear in a new way to the individual; and extrapolation, which involves prediction of a given happening, based on understanding the information that the individual already has. Last is application, in which the information conveyed to the individual is actually applied to the situation in which the learning process took place, in this case, practices of caring for children.

An example of a question from the knowledge area is "At which age is it recommended to start weaning a child?" In this case, the mother may just remember the age, without understanding why weaning is recommended then, and without doing it. An example of a translation question is "What do you think it means if your child's weight on the growth chart is between the two solid lines?" An example of an interpretation question is "By looking at the growth chart, how do you know if your child is putting on weight?" An example of an extrapolation question is "If your child's weight was below the lower (i.e., dotted) line, what could you do to help him gain weight?" An example of an application question is "Explain what you see in your child's growth chart."


Evaluating the understanding and use of the growth chart

This evaluation followed a pre-intervention post-intervention design. The basic evaluation was done after the field auxiliaries had been visiting the mothers for at least four months to weigh their children. The second evaluation took place one year after monthly visits by the fieldworkers had begun.

At each evaluation mothers received a questionnaire containing 14 questions designed to cover the first two cognitive areas. The same schedule was used for the pre-evaluation and the post-evaluation. The questionnaire was administered by a field psychologist during home interviews.

Information on application for the basic evaluation was collected through specific questions. For the final evaluation, a field anthropologist visited mothers in their homes to observe what they were actually doing in relation to the activities promoted by the health auxiliaries. He reported his observations on a specific interview form that covered the same topics as those of the fieldworkers on their monthly visits.

Twenty items on the questionnaire assessed mothers' attitudes towards the growth chart, the health auxiliaries, and weighing their children. The answer to each question was yes, sometimes, or no, reflecting a positive, undetermined, or negative attitude to the specific question.


Statistical analyses

The statistical significance of the difference between the mother's scores in the pre-evaluation and post-evaluation questionnaires was assessed by the c2 statistic [11]. The significance of the child's nutritional status at the end of the protect i.e.. whether the child mostly gained weight, mostly maintained weight, or mostly lost weight—was assessed by the Kruskal-Wallis test. Correlations among knowledge, understanding, and use scores and the child's nutritional status at the end of the project were established by Spearman's rank correlation coefficient. The mother's performance of the different activities, evaluated as part of the intervention according to her literacy, was evaluated by the Kolmogorov-Smirnov test.


Of 100 eligible mothers, 85 agreed to participate in the project. During the first months of the programme, three more mothers, each from a different community, asked to be included in the growth-monitoring project, so the final study population consisted of 88 mothers. Eighty percent of them were literate, and 71% were under 30 years of age. Their performance in the learning process at the time of the pre-evaluation and post-evaluation measures is shown in table 1. All the areas evaluated showed an increase between the first and the second evaluations, with significant changes (p < .05) in knowledge, interpretation, and application. The first and last areas showed a similar magnitude of change (30%). Mothers ranked higher in knowledge than in any other area. Ninety-six percent had a positive attitude towards weighing their children from the outset. Eighty-nine percent had a positive attitude towards the growth chart and 94% had a positive attitude towards the health auxiliaries at the beginning. Over the course of the study these percentages increased to 96% and 98%, respectively (p < .05).

TABLE 1. Mothers, performance related to Bloom's taxonomy of the learning process on basal and final evaluations

Stage of evaluation Knowledge Understanding Application
Translation Interpretation Extrapolation
Basal 58a 62 58a 72 44a
Final 88a 79 77a 76 74a

Numbers represent grades on a scale of zero to 100.
a. p<.05.


At the beginning of the project, according to the Gomez criteria [1], using the fiftieth percentile of National Center for Health Statistics (NCHS) [12] growth curves as the reference population, most of the children had some degree of malnutrition: 46% had first-degree and 16% had second-degree malnutrition (see FIG. 2. Mean weights of study children compared with NCHS 50th percentile). Only 38% were well nourished. Although all children except one had at least one period of weight loss during the study, at the end of the project 48% were well nourished (50% had recovered from different degrees of malnutrition), 36% had first-degree malnutrition, and 15% had second-degree malnutrition. These changes in nutrition status were highly significant (p < .0001).

During the study, 81% of the children mostly gained weight, 7% mostly maintained weight, and 12% mostly lost weight. The children who were well-nourished at the end of the study were predominantly (86%) those who gained weight. Seventy-five percent and 77%, respectively, of children with first- or second-degree malnutrition at the end of the study were gainers (table 2). These differences were not statistically significant.

TABLE 2. Degree of malnutrition at the end of the intervention in relation to the predominant weight change of children during the intervention (in two-thirds of observations)

Degree of malnutrition % of children who predominantly:
Lost weight Maintained weight Gained weight
None 9.3 4.7 86.1
First-degree 15.6 9.4 75.0
Second-degree 15.4 7.7 76.9


There was a significant tendency for older mothers to have children who were gainers (p < .01). Literate mothers were also more likely than illiterate mothers to have children who were gainers (p < .01), and their higher knowledge and understanding scores, as well as each of the application scores, were related to a weight gain trend (p < .01).

Most mothers complied with the health care activities promoted by the auxiliaries, except for boiling drinking water, which was performed by only 18% of the mothers. All of the activities were significantly (p < .001) more likely to be performed by mothers of children who predominantly gained weight during the project than by mothers of children who predominantly maintained or lost weight (table 3). There was no significant association between mothers' literacy status and their performance of the activities.

TABLE 3. Predominant weight change of children during the intervention (in two-thirds of observations) in relation to the percentage of mothers who performed different activities promoted by health auxiliaries

Activity N % of mothers who performed the activity for a child who predominantly:
Gained weight Maintained weight Lost weight
Helps undress child before weighing 75 89.6 4.1 6.2
Helps place child on the scale 75 78.6 7.1 14.3
Gives food according to child's age 75 82.2 5.5 12.3
Feeds appropriate food to child age 6 mo 75 82.2 5.5 12.3
Boils drinking water 71 92.3 0 7.7
Keeps cookware clean 60 83.6 3.6 12.7
Cleans and trims child's nails 75 79.6 7.4 12.9
Has adequate sanitary habits 41 85.3 2.9 11.8
Knows about child's vaccination scheme 74 81.3 6.3 12.5

All of the activities were significantly (p < .001) more likely to be performed by mothers of children who predominantly gained weight during the project than by mothers of children who predominantly maintained or lost weight.


The study area shares several characteristics with other areas in the highlands of central Mexico, which gives external validity to our findings. Such is the case for the distribution of mothers according to literacy status and age. Consistent with recent trends in women's education in Mexico, literate mothers were clearly predominant in the younger age groups. Furthermore, the percentage of malnourished children in the area is in close agreement with the general status in the rest of the country [13], that is, a large proportion of the population suffers from marginal malnutrition, although the severe forms (kwashiorkor, marasmus) are rare.

The study was not meant to evaluate the impact of the introduction of the growth chart in improving children's growth. We considered that the international literature contained enough evidence to support growth monitoring as an efficient way to follow children's growth and survival in the developing world [4, 5, 14 16], and our results supported this concept. Thus, we were more concerned with finding an effective way to use growth charts at the community level than with evaluating their effectiveness.

The mothers had lower scores in application than in either knowledge or understanding, particularly in the basal evaluation. This was to be expected, as it is easier to remember by heart a new piece of information than it is to interpret and use it. However, it was encouraging to find that the mothers had higher scores in all the areas evaluated at the time of the final evaluation, as this can be interpreted as a successful result of the programme.

Whereas most mothers had a positive attitude towards the different components of the programme, the highest positive attitudes were those directed to weighing a child. The positive attitudes in this area may be due to better understanding of a specific action, such as weighing the child, compared with something further removed from their everyday activities, such as the growth chart. This may be interpreted as reflecting a better grasp of a pragmatic activity than of a concept.

The comparison of mothers' attitudes towards the different components of the project in the basal and final evaluations points to another important indicator of the success of the intervention. Post-intervention, there was a significant decrease in the number of mothers who had a negative attitude towards the growth chart and towards the health auxiliaries. It should also be noted that some mothers who were not originally recruited asked to be included in the programme.

The high percentage of positive attitudes towards the health auxiliaries reflected the rapport that they established with the mothers. Some children, after a year of being weighed each month, readily recognized the health auxiliary and willingly slipped into the weighing pants, or found their growth chart and brought it themselves to the auxiliary so that she could write down their weight.

In several cases, a common activity of mothers, cooking and storing tortillas (the most common way to eat maize in Mexico) on top of each other as they come out of the cooking pan, was compared with the accumulation of bubbles in the growth chart. The health auxiliaries explained that, for children to put on weight, mothers had to "put in" tortillas (food), which would amass in the form of weight gain, reflected in an increase of bubbles (grams) in the growth chart.

At present, the growth-monitoring activities have been incorporated into a larger-scale surveillance system, using the bubble chart in 35 communities in the area, with several hundred children being weighed each month. The information gathered is periodically updated in a computer-based information system, as part of a larger-scale programme, in which identification of a problem is immediately followed by an activity meant to solve the problem. These activities are part of operations research and will be addressed in a different publication.


  1. Gómez F. Desnutrición. Bol Med Hosp Inf Mex 1946; 3:543.
  2. Waterlow J. Classification and definition of protein-calorie malnutrition. Br Med J 1972;3:566.
  3. Gopalan C. Growth monitoring. Intermediate technology or expensive luxury? Lancet 1985;2:1337-42.
  4. Gopalan C, Chatterjee M. Use of growth charts for planning child nutrition. A review of global experiences. Series 2. New Delhi: Nutrition Foundation of India, 1985.
  5. George S. Latham M, Ethirajan N. Evaluation of effectiveness of good growth monitoring in south Indian villages. Lancet 1993;342:348-52.
  6. Morley D, Woodland M. See how they grow. Hong Kong: Macmillan Education, 1979.
  7. Griffiths M. Growth monitoring. Practical considerations for primary health care projects. Washington, DC: World Federation of Public Health Associations, 1985.
  8. Griffiths M, Berg A. The bubble chart: an update on its development. Food Nutr Bull 1988;10(3):71-4.
  9. Allen L, Pelto G. Chávez A. The collaborative research and support program on food intake and human function: Mexico project. Storrs, Conn, USA: University of Connecticut, Department of Nutritional Sciences, 1987.
  10. Bloom B. Taxonomía de los objetivos de la educación. Buenos Aires: Editorial Ateneo, 1971.
  11. Mascie Taylor C. Analysing cross-sectional anthropometric data. Eur J Clin Nutr 1994;48:S190-201.
  12. Hamill P. Dritz T. Johnson C, Reed R. Roche A. NCHS growth curves for children birth-18 years. Vital and health statistics, series 11, no. 15. Washington, DC: Department of Health, Education and Welfare, 1977.
  13. Sepúlveda AJ, Lezana MA, Tapia CR, Valdespino JL, Madrigal H. Kumate J. Estado nutricional de preescolares y mujeres en Mexico. Resultados de una encuesta probabilistica nacional. Gac Med Mex 1990;126:207-25.
  14. Griffiths M. Growth monitoring: making it a tool for education. Ind J Pediatr 1988;55:S56-66.
  15. Shekar M, Latham M. Growth monitoring can and does work: an example from the Tamil Nadu integrated nutrition project in rural South India. Ind J Pediatr 1992; 59:595-8.
  16. Bredow M, Jackson A. Community based, effective low cost approach to treatment of severe malnutrition in rural Jamaica. Arch Dis Child 1994; 71:997-303.

Contents - Previous - Next