Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Growth monitoring in the context of a primary health care programme

Fazlul Karim, Nasreen Huq, Laurine Brown, and A. Mushtaque R. Chowdhury

The study

Objectives

The objectives of the study were as follows:

• to determine the level of operational efficiency of the BRAC approach to growth monitoring;
• to evaluate the effectiveness of monitoring in increasing the use of other child-survival technologies such as oral rehydration therapy, immunization, and vitamin A capsules;
• to measure the changes in the nutrition status of the children due to the monitoring;
• to assess the perceptions of the villagers about the monitoring.

Methods

To address the above objectives, data from primary as well as secondary sources were used.

For the first objective, data were collected from eight villages of Manikganj Sadar and Gheor thanas. In these villages the Research and Evaluation Division of BRAC had been carrying out a passive surveillance system through which, among other things, births and deaths had been recorded since 1987. Thus the precise ages of children born since 1987 were available. As the target population for the growth monitoring was children under two years of age, households with at least one child of this age were chosen for study.

Data covering different aspects of the operational efficiency of the monitoring were collected by a variety of methods. For example, the extent of the programme's coverage of target children was determined by asking households whether they had growth-monitoring cards or not; the accuracy of the records of the children's ages was determined using registration data; the accuracy of weighing scales was measured using a standard 5-kg weight on eight round (25-kg) and six cylindrical (15-kg) Salter scales; the consumption of supplementary diet was investigated through 24-hour recall; and the accuracy of recorded weights was checked by direct observation.

To determine mothers' knowledge and how they interpreted the growth card, the following questions were asked: (1) On which part of the card should your baby's weight be for you to know that he or she is growing well? (2) Which part is bad? (3) What happens if the weight mark falls below the green belt? (4) What happens if the weight mark is in the green belt? (5) What happens if the weight mark is above the green belt? (6) What does it mean if the present month's weight is more than the previous month's? (7) What if the present month's weight is less than the previous month? All these data for the eight villages were collected during November-December 1990.

To ascertain whether the growth monitoring increased the use of other child-survival strategies, data were used from a survey of 3,048 households in five of the six primary health care thanas in September 1989 to assess the midcourse effects of the BRAC programme on the knowledge and use of its components. The survey asked mothers with children under two years of age whether, at the time of the survey, they were participating in the programme or had ever done so. Since the survey also collected data from the same mothers on the use of other child-survival practices promoted by BRAC, this facilitated analysis of the data to see if participation in growth monitoring was related to their use of those components. No anthropometric data were collected during this survey.

To measure the impact of the monitoring on the children's nutrition status, data were used from a nutrition surveillance project in which BRAC had been participating since early 1990, together with several other NGOs, by providing data from two of its primary health care thanas (Saturia and Santhia). Through this project, the growth of a large number of children 6-59 months old was monitored quarterly using several anthropometric measures, under the supervision of Helen Keller International [7]. For the present study, all 496 children 623 months old from the two BRAC areas covered by the second and the third rounds of the surveillance were revisited. They were grouped according to whether or not they were participants in BRAC's growth monitoring, and the nutrition status (weight for age) of the children who participated was compared with that of those who did not participate.

To elicit villagers' perceptions of and attitudes toward the monitoring, qualitative data were collected through participant observations and 14 focus-group discussions [6].

Findings

Accuracy of age determination

Comparison of the children's ages recorded by health workers for growth monitoring with those recorded by the sample registration system operated by BRAC showed that in 22.5% of the cases the match was perfect (table 1). In 69% the difference between the two sources was 30 days or less.

TABLE 1. Discrepancies between the ages of children 0-23 months old shown in growh-monitoring records and their actual ages determined from birth registration records

Data from a special survey in eight villages in two thanas.

Accuracy of weighing

Seven of the eight round scales checked gave exact weights. Only one of the six cylinder scales gave exact weights, while the other five gave weights in excess of 5 kg. In 89% of cases the weights were correctly read and recorded by the health workers.

In 80% of the cases the scale was adjusted to zero at the start of the weighing session. This was done infrequently afterwards, however; in only 47 % of the cases was the scale adjusted at least once before the end of the day.

Coverage of target children

Our survey indicated that 43% of the target children under two years of age were actually covered by the monitoring, and the coverage for those under six months was only 12% (table 2). The 1989 survey of five thanas found that only 35.7% of households with children under two years of age had growth cards (ranging from 24% in Rangpur to 44% in Saturia). According to the nutrition surveillance project data for two thanas, 38% of the children 6-23 months old had growth cards (51% in Saturia and 26% in Santhia thana); there was no significant difference in the number of boys and girls participating (table 3).

TABLE 2. Coverage of children 0-23 months old in the growth-monitoring programme by age group

Data from a special survey in eight villages in two thanas.

TABLE 3. Coverage of children 6-23 months old in the growth-monitoring programme by sex

Data for two thanas from the Nutritional Surveillance Project conducted under the auspices of Helen Keller International (rounds II and III).

Mothers' understanding of the growth-monitoring card

Forty-three per cent of the mothers who participated in the monitoring answered all seven questions about the card correctly, compared with 8% of non-participants. Among the participant mothers, 12% could not answer any of the questions correctly, compared with 45% of the non-participants.

Time spent by mothers in growth-monitoring sessions

The average total time spent by a mother from registration to completion of health education was 16.4 minutes. Average times spent weighing and recording and in receiving nutrition education were 1.7 and 5.3 minutes respectively. The time spent travelling to and from the centre was not considered.

Community participation

Community participation took a number of forms. As mentioned, the community provided various resources for the smooth conduct of the sessions. The community also provided volunteers, whose role increased as the programme matured. Depending on how long a growth-monitoring centre had been present in a village, their number varied from none at the initial stage to three in final months.

Most of the volunteers (85%) were women, with an average age of 42 years. The majority were directly or indirectly connected with other BRAC activities as village health workers, trained TBAs, or village health committee members. Approximately half of them were literate. Forty-seven per cent of the volunteers from Manikganj and 17% from Gheor were landless. (In Manikganj BRAC has concentrated its development programmes only on landless people.)

Impact on other child-survival practices

In evaluating the effect of participation in growth monitoring on other practices contributing to child survival promoted by BRAC, households from which at least one child in the target age range participated in the programme were designated participants, and those from which no child in the age group participated were designated non-participants. The participant households performed significantly better than the non-participant group in regard to all of the practices surveyed but one (waste disposal in a designated place; table 4).

TABLE 4. Proportions of households or individuals observing various child-survival practices according to household participation in growth monitoring

Karim and Chowdhury, unpublished data, 1991).
a. Oral rehydration solution of salt and unrefined sugar in drinking water, given to children experiencing diarrhoea.

Table 5 shows the mean frequencies of feeding different items for participants and non-participants. No significant differences were found between the groups.

TABLE 5. Mean frequency of intake of various foods by children 13-24 months old during the 24 hours preceding the survey

Data from a special survey in eight villages in two thanas.
Values are means ±SD.
a. N=63. b. N=49.

We also compared selected background characteristics of participants and non-participants. The two groups were largely identical with respect to landholding, occupation of the father, and literacy of the mother (table 6).

TABLE 6. Selected characteristics of participant and non participant households

Karim and Chowdhury, unpublished data, 1990.

Nutrition status

Nutrition status was evaluated using the Gomez criteria for weight for age (>90% of the median, normal; 90%75%, mild malnutrition; 75%-60%, moderate malnutrition; < 60%, severe malnutrition). Table 7 shows that a somewhat higher proportion of the growth-monitoring participants than of the non-participants had normal weight for age, 10.6% and 5.8% respectively, but this difference was not statistically significant (p=.051). The proportion who were severely malnourished appears to be higher among the non-participants, but the cell frequencies are unfortunately too small to support any conclusion.

TABLE 7. Nutrition status of participant and non-participant children 6-23 months old

Calculated from data of the Helen Keller International Nutritional Surveillance Project (rounds II and III) and a follow-up survey by BRAC.

Perceptions of growth monitoring

Non-participating mothers said that they had heard about growth monitoring but nobody had asked them to weigh their babies. They further indicated that they were not aware of the usefulness of weighing babies, since the health workers had not explicitly explained it to them.

The majority of the participating mothers believed that the monthly monitoring sessions informed them of their children's growth status and allowed them to take necessary measures to improve the status in a timely fashion. The sessions also enabled them to receive health education.

A number of mothers initially faced resistance to participating, as some family members thought that it was not good for the baby ("brings sickness"). Gradually such obstacles were overcome through motivation by BRAC workers.

The village leaders were aware of the sessions and were generally supportive and convinced of their usefulness.

Contents - Previous - Next