Contents - Previous - Next

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

Nutrition and health

Maternal malnutrition, outcome of pregnancy, and a simple tool to identify women at risk

Y. K. Husaini, M. A. Husaini, Z. Sulaiman, A. B. Jahari, Barizi, S. T. Hudono, and D. Karyadi
Ministry of Health, National Institute for Health Research and Development, Bogor, Indonesia


Infants who weigh less than 2,500 g represent about 14 per cent of all live births in Indonesia each year [9]. Although a number of pre-natal influences affect this low birth weight, one of the most important causes is maternal malnutrition, particularly chronic calorie deficiency [6, 7]. In view of the risks of early death or permanent disability associated with low birth weight, the government wishes to control this problem and decrease the incidence to 7 per cent by the year 2000 [9].

Therefore, an appropriate technology that can be used to identify women at risk is urgently needed. The technology proposed is a mother's health card that can be used to evaluate the intervention and also to increase the effectiveness of programmes in promoting health and improving the nutritional status of pregnant women.

Functional consequences of maternal malnutrition

Rosso and Cramoy reviewed literature on weight change during pregnancy associated with foetal growth [11]. In various mammalian species in which maternal malnutrition has been induced during pregnancy, the reduction in foetal weight tends to be proportionate to the loss of maternal body weight [2, 11, 19].

The consequences of maternal malnutrition for the foetus depend on the timing, severity, and duration of the maternal dietary restriction. The possible harmful effect observed in experimental animals is a deficit in brain growth associated with maternal malnutrition [14]. It has also been observed that the lower the birth weight of term infants, the greater the chance of a subsequent deficit in mental capacity [4, 8, 17, 18]. In general, birth weight, as a reflection of intra-uterine growth, is a determinant of the child's potential for survival and future physical and mental health [16].


The study population consisted of 2,457 pregnant women residents of the city of Bogor (West Java, Indonesia) and its suburbs. Most of the women came from socio-economic classes ranging from middle to lower, and represented a social, economic, and educational stratum that usually receives the services of community health centres. The husbands' jobs included government employee, private worker, vendor, and farmer.

The women in the study population were followed starting from the eighth to the sixteenth week of pregnancy (calculated from the first day of the last menstruation) for periods extending up to delivery. Of these 2,457 women, 1,125 were excluded from analysis because: (a) the number of weighings was not large enough to provide the required data with sufficient accuracy; (b) the pregnant woman moved to her parent's home in another area; (c) the birth weight was not recorded; (d) the woman underwent an abortion; or le) the woman's baby was stillborn.

Measurements, consisting of height, weight, arm circumference, and triceps skinfold thickness, were obtained for all pregnant women. Weight was recorded with a Detecto scale, arm circumference with a plastic-coated tape, and triceps skinfold thickness with a Harpenden calliper. All measurements were made using techniques similar to those described by Jelliffe [5]; measurements were taken at the first ante-natal visit and every four to six weeks afterwards, except for height, which was recorded only at the first visit. Measurements for most women were then taken until the thirty-sixth to fortieth week of pregnancy. Since many women were not measured during earlier weeks of pregnancy, data for the first seven weeks were too sparse to have any representative value.

The infant data collected included methods of delivery and weight. Measurements of the infants' birth weight were made by midwives or skilled research assistants between 3 and 24 hours after the delivery on a beam balance scale that could be read to the nearest 10 g.

The number of cases in each period varied because: not all women reported at the same time in the early weeks of pregnancy, not all weighings were regular; and some women were delivered too early.

Information on age, education, parity, history of the previous delivery, chronic diseases, and interval since last pregnancy were recorded at the first visit. Other information was obtained in each trimester on dietary intake (using the method of 24-hour recall twice), haemoglobin concentration (by the cyanmethemoglobin method), protein in urine, oedema, and blood pressure.

Data Analysis

Data analyses included only mothers with singleton, live-born babies, whose gestation lasted from 36 to 40 weeks, during which four or more weight measurements were recorded. Excluded from the analyses were women with any of the non-nutritional disorders that can affect foetal growth or maternal weight, such as maternal diabetic mellitus, pre-eclampsia, and eclampsia.

Correlations, as well as simple and multiple regression analysis, were used to establish which variables had a significant effect on maternal weight during pregnancy and on the birth weight of the infants. Chi-square analysis was used to evaluate grouped data.

Women were grouped into two categories according to the birth weight of their infants: group 1 comprised women who delivered low-birth-weight infants (<2,500 g) and group 2 those who delivered adequate-birth-weight infants (>2,500 g). Average curves were drawn for groups 1 and 2 to evaluate the maternal weight trends at different stages of pregnancy for certain maternal heights. Next, analyses were undertaken to determine the cut-off points between the two groups by sensitivity and specificity tests. Finally, the cut-off lines for each height were made in smooth curves by using overlapping regressions.


The mean age of the women in the sample at the first visit was 25 years, with a range of 14 to 44 years. Twenty per cent (n=260) of the women were nulliparous, and 13.4 per cent (n=179) were grandiparous (equal to and above six parities). The mean height of the women was 149.5 cm ± 4.97 cm (SD). Of the 1,332 women in the study, 1.4 per cent (n=19) had heights below 140 cm, and 3.2 per cent (n=42) had heights equal to or above 160 cm. Six per cent (n=82) of the women in the study had no education, and 19.0 per cent (n=253) had one to three years of primary school. Thus, 25.2 per cent (n=335) of the women had no education or less than four years of primary school. Only 0.4 per cent (n=5) had graduated from a university with a B.Sc. degree.

TABLE 1. Risk factors in pregnancy associated with low birth weight

Risk factors Birth weight Chi2 P
<2,500 g (n=138) >2,500 g (n=1,194)
N Percentage N Percentage
1. Age >40 years 12 8.7 45 3.8 7.2 P<0.01
2. Height <140 cm 6 4.3 13 1.1 9.8 P<0.01
3. No education or one to three years of primary school 56 40.6 279 23.4 6.5 P<0.05
4. A history of chronic diseases in the mother 14 10.1 66 5.5 5.5 P<0.05
5. Interval since last pregancy <6 months 9 6.5 32 2.7 6.3 P<0.05
6. Oedeme, urine positive for Protein, end high blood pressure 19 13.8 80 6.7 8.9 P<0.01
7. Hb <8 g/dl 20 14.5 84 7.0 9.4 P<0.01

Medians of arm circumference and triceps skinfold thickness at different stages of pregnancy

The mean calorie intake, when compared to the national estimated average energy requirement (AER), was 70 per cent (1,540 kcal) in the first trimester, 80 per cent (1,760 kcal) in the second trimester, and 80 per cent (1,760 kcal) in the third trimester. The mean protein intake was 85 per cent of the national recommended dietary allowance (RDA) (41 9) in the first trimester, 100 per cent (48 9) in the second trimester, and 100 per cent (48 9) in the third trimester of pregnancy. The consumption of calories was low, but the consumption of protein was nearly adequate.

The delivery types of women in the study were found to be: 83.8 per cent (n= 1,678) spontaneous delivery; 1.6 per cent (n=32) still birth; 1.3 per cent (n=26) abortion; and 13.3 per cent (n=266) not reported. Of the 1,332 women included in the final analysis, 10.4 per cent (n=138) of the infants weighed less than 2,500 g at birth (low birth weight).

The risk factors found to have been significantly associated with low birth weight were: (a) age over 40 years; (b) height less than 140 cm; (c) no formal education or less than four years of primary school; (d) a history of chronic infectious diseases, (e) an interval since the last pregnancy equal to or less than six months; (f) a combination of oedema, urine positive for protein, and high blood pressure (either systolic blood pressure above 140 mm Hg or diastolic reading above 90 mm Hg); and (g) haemoglobin concentration below 8 g/dl. The details of those risk factors associated with low birth weight are presented in table 1.

The triceps skinfold thickness and arm circumference data are shown in figure 1. Neither the median for triceps skinfold thickness nor arm circumference showed any significant increments when plotted against the number of weeks of pregnancy. These measurements, therefore, could not be used to evaluate foetal growth and the nutritional status of the women during pregnancy.

The maternal weight data are shown in figure 2. A trend of increase in the median maternal weight when plotted against increments of weeks of pregnancy is clearly demonstrated. The more weeks of pregnancy, the heavier the maternal weight. Therefore, in this study, it would seem that the actual pregnancy weight is much more useful in identifying women who may deliver low-birth-weight infants than triceps skinfold thickness or arm circumference measurements.

This study also examined the correlation between maternal weight during pregnancy and maternal height, parity, maternal age, haemoglobin, and blood pressure. The correlation between maternal weight and maternal height was highly significant (r2=0.28, n=1,303). On the other hand, the correlations between maternal weight and each of the other factors-for example, between maternal weight and maternal age (r2=0.02, n=1,310) and between maternal weight and parity (r2=0.007, n=1,052)-were very low. These findings support the idea that maternal height and weight during pregnancy have effects on the birth weight of the infants.

The next steps were: (i) to determine the regression pattern for weights of mothers who deliver low-birth-weight infants; (ii) to find the regression pattern for weights of mothers who deliver adequate-birth-weight infants; and (iii) to carry out sensitivity and specificity tests at different weeks of pregnancy.

Medians of maternal weight at different stages of pregnancy

Figure 3 shows the cut-off lines between the mothers who delivered low-birth-weight infants and those who delivered adequate-birth-weight infants. On the basis of our findings a mother's health card was designed to provide pregnant women with information about their nutritional status and whether they are at risk during pregnancy.

Guidelines for using the mother's health card

On the mother's health card health personnel enter the name of the community health centre, the name of the pregnant woman, and her address, age, formal education, husband's name and job, the first day of her last menstruation, and approximate date of delivery. Information concerning the woman's previous obstetric performance is entered in a tabular form. The woman's weight and height are then measured by the health personnel.

The health personnel record such observations as foetal position, foetal heart rate, maternal blood pressure, oedema, immunization status against tetanus, iron supplementation during pregnancy, and haemoglobin concentration. Information concerning risk factors is also solicited. The health personnel will complete the data whether the woman is at risk or adequately healthy. In the at-risk table, any of the following risk indicators, if present, should be mentioned: (a) height if less than 140 cm; (b) age if less than 20 years; (c) age if equal to or above 35 years, (d) lack of formal education or less than four years at primary school; (e) interval since the last pregnancy if equal to or less than six months; (f) presence of chronic infectious diseases; (g) diastolic blood pressure above 90 mm Hg; and (h) severely anaemic or having a haemoglobin concentration below 8 g/dl.

The weight is plotted on the graph against the weeks of pregnancy. The health worker thickens the appropriate available curve for a woman's height in the graph with a coloured pencil or pen. A woman whose weight plot is above the appropriate height curve is likely to deliver a baby of adequate birth weight while one whose weight plot is below the appropriate height curve is likely to deliver a low-birth-weight infant.

For the next ante-natal visit, the weight is again plotted on the graph against weeks of pregnancy. A line is then drawn between the point representing the previous weight and

1. Although in Indonesia the week of pregnancy is usually easy to calculate from information on the last menstrual period, it should be noted that errors in assigning the week of pregnancy have practically no effect on the nature and use of the method. The slope of the weight curve is relatively independent of the week of pregnancy. the current weight. If the line is above the appropriate height curve, it means that the nutritional status of that woman is likely to be adequate; if the line is below the height curve, the woman needs special attention or an intervention to prevent her from delivering a low-birthweight infant.

Messages for nutrition education are also provided to promote better maternal and foetal nutrition. The pregnant women are persuaded to eat larger quantities and a greater variety of the foods locally available.

Mother's health card for national health and nutrition programmes

Low birth weight has been reported to be a serious threat to survival in the newborn period and to normal physical and mental development in the post-natal period [3, 15]. In babies with birth weights under 1,500 g, pre-natal complications result in brain retardation, which detracts from their later performance. Their average IQs are lower than those of the general population, and they do poorly at school [ 11 ]. For infants whose birth weights are closer to normal, development will depend on post-natal nutrition as well as on the physical and social environment [15].

The incidence of low birth weight can be reduced if pregnant women at risk are identified and steps taken to reduce the risk. For this approach the women need to be identified early in pregnancy, so that they themselves, as well as health personnel (e.g. midwives, nutritionists, and physicians), can be alerted to their condition [12, 13]. To achieve this goal, the mother's health card-which is simple, does not require extensive equipment and training, and can be used by workers at the community level-has been introduced.

Curves for the assessment of weight during pregnancy, in relation to height

The objective of the mother's health card is to promote better maternal and foetal nutrition by increasing the level of awareness of the family and health worker through the graphic presentation of simple body measurements: height and weight. The card can be used to monitor and evaluate the nutritional status of women during pregnancy; it is also a health education tool that encourages women and their families to participate in self-care and gives a persuasive reason for them to use the health-care services available in the community.

This mother's health card has been accepted for use in the Mother and Child Health and Family Nutrition Improvement Programmes in Indonesia, following a one-year validation study in three provinces [10]. It is hoped that other countries will also explore the use of this procedure.


1. G. H. Beaton, and J. M. Bengoa, "Practical Population Indicators of Health and Nutrition," in Nutrition in Preventive Medicine, WHO Monograph Series, no. 62 (WHO, Geneva, 1976).

2. B. N. Berg, "Dietary Restriction and Reproduction in the Rat," J. Nutr., 187:344 (1965).

3. J. E. Brown, H. N. Jacobson, L. H. Askue, and M. G. Peick, "Influence of Pregnancy Weight Gain on the Size of the Infants Born to Underweight Women," Am. J. Obstot Gynocol., 57: 13 (1981).

4. C. M. Driellien, "The Small-for-Date infant: Etiology and Prognosis," Pediatr. Olin. North Am., 17: 9 (1970).

5. D. B. Jelliffe, The Assessment of the Nutritional Status of the Community (WHO Geneva)

6. A. Lechtig, H. Delgado, R. Lasky, C. Yarbrough, R. E. Klein, J. P. Habicht, and M. Behar, "Maternal Nutrition and Fetal Growth in Developing Countries," Am. J. Dis. Child., 129: 563 (1975).

7. S. Martoatmodjo, Muhilal, M. Enoch, M. A. Husaini, C. W. Angkuw, and Prawiranegara," Anemi gizi pada wanita hamil di desa Bendunga, Kabupaten Bógor," in S. Martoatmodjo, I. Tarwotjo, Djumadias Abunain, and D. Karyadi, eds., Penelitian Gizi dan Makanan (Balai Penelitian Gizi Unit Semboja, Bogor, Indonesia, 1972),p. 11.

8. H. C. Miller, and K. Hassanein, "Fetal Malnutrition in White Newborn Infants: Maternal Factors," J. Pediat, 52: 504 (1973).

9. Ministry of Health, Republic of Indonesia, National Health System, 2nd en. (Ministry of Health, Jakarta, Indonesia, 1982).

10. Ministry of Health, Republic of Indonesia, National Workshop on Food and Nutrition, Bogor, Indonesia, 14-17 July 1978, Proceedings.

11. P. Rosso and C. Cramoy, "Nutrition in Pregnancy," in M. Winick, ed., Human Nutrition: A Comprehensive Treatise. Nutrition Pre- and Postnatal Development, vol. I (Plenum Press, New York, 1979), p. 133.

12. P. Rosso and C. Cramoy, "Nutrition in Pregnancy", in Pregnancy," Am. J. Clin. Nutr., 41: 644 (1985).

13. K. P. Shah, "Surveillance Card for Married Women for Better Obstetric Performance," J. Obstot. Gynaecol. India, 28: 1015(1978).

14. Z. A. Stein and M. Susser, "Mental Retardation," chapter 38 in John M. Last, ed., Maxcy-Rosenau, Public Health and Preventive Medicine, 11th ed. (Appleton Century-Crofts, New York, 1980), pp. 1276-1277.

15. P. S. Venkatachalam, "Maternal Nutritional Status and its Effect on the Newborn," Bull. Weld. Health. Org., 26: 193 (1962).

16. J. Vermeersch, "Maternal Nutrition and the Outcome of Pregnancy," in B. S. Worthington-Roberts, J. Vermeersch, and S. R. Williams, ads., Nutrition in Pregnancy and Lactation, 2nd ad. (C. V. Mosby Company, St. Louis, 1981), p. 13.

17. G. Wiener, "The Relationship of Birth Weight and Length of Gestation to Intellectual Development at Ages 8 to 10 Years," J. Pediatry., 76: 694 (1970).

18. M. Winick and P. Rosso, "Malnutrition and Central Nervous System Development," in J. W. Prescott, M. S. Read, and D. B. Coursin, ads., Brain Function and Malnutrition: Neuropsychological Methods of Assessment (John Wiley & Sons, New York, 1975).

19. M. Young and E. M.Widdowson, "The Influence of Diets Deficient in Energy, or in Protein on Conceptus Weight, and the Placental Transfer of a Non-metabolisable Amino Acid in the Guinea Pig, " Biol. Neonate, 27: 184 (1975).

Contents - Previous - Next