Indian Journal of Community Medicine

ISSN No. 0970-0218

BT Rao, Arun Kumar Aggarwal, Rajesh Kumar

Abstract

Objectives: (1) To assess the magnitude of the problem of low birth weight (LBW) in a rural area of Haryana (2) To study the association of caloric and protein intake in third trimester of pregnancy with birth weight. Study Design: Longitudinal study. Sample Size: One hundred and forty pregnant women. Study Area: Ten purposively selected villages in the rural area of Naraingarh block in Haryana. Study Tool and Data Collection: Pre-tested questionnaire was administered to record information regarding socioeconomic status, antenatal care, nature of physical activity and dietary intake in 24 h between September 2001 and August 2002. Data Analysis: For categorical variables, Chi-square test was used, and for numerical variables, t-test was used. Multivariate analysis was done for variables that were significant in bivariate analysis. Results: The prevalence of low birth weight (less than 2500 g) was 24.3% (95% CI 17.4-32.2%). The mean caloric intake during three dietary assessments was 1695 ± 182.8 kcal. The mean protein intake during three dietary assessments was 50.8 ± 9.27 g. The higher prevalence of LBW babies was observed in pregnant women with mean caloric intake of less than 1500 kcal and mean protein intake of less than 40 g (P < 0.001). In multivariate analysis, the association of caloric intake (P < 0.01) and weight of the mother (P = 0.02) was independently associated with LBW. Conclusions: Low caloric intake in the third trimester of pregnancy and maternal weight are significantly associated with the birth weight of babies.

Keywords: Caloric intake, low birth weight, protein intake, pregnancy

Low birth weight (LBW) remains an unresolved important national concern in India. Twenty-nine percent of infant mortality rate is associated with LBW in India.¹ Twentythree percent of the newborns in India have LBW. The prevalence is slightly higher in rural areas (24.1%) than in urban areas (21%).² The prevalence has remained almost static over the last one decade. In a rural area of Haryana, LBW prevalence was 25.3% in 1982-1984³ and 25% in 1997-1998.⁴ Nutrition intake in pregnancy is among one of the many factors associated with LBW in developing countries.⁵ Nutritional needs increase during pregnancy, especially in the second and third trimesters of pregnancy.⁶ Nutritional counselling to mothers early in pregnancy can help improve dietary intake during pregnancy. The high level of TT administration to pregnant women in rural Haryana indicates that antenatal mothers do come in contact with health services.² However, contact with health services mostly occurs in the third trimester. Thus, the third trimester of pregnancy is the only period where nutritional counselling can be done. The present study was planned to assess the magnitude of the problem of LBW in a rural area of Haryana and to study the association of birth weight with caloric and protein intake in the third trimester of pregnancy.

Materials and Methods

The present study was carried out in the rural areas of district Ambala, Haryana. The villages are well connected by road and have water, electricity and telephone connections. A sub-district hospital is at about 20-30 min of motorable distance from these villages. The study was initiated in September 2001. A prospective longitudinal study design was followed. A sample size of 140 antenatal women was chosen considering the LBW prevalence of 25%, desired precision of LBW of 7%, and design effect of one and alpha of 0.05. It was estimated that at a birth rate of 25 per 1000 population, there would be about 12-13 antenatal women in the second and third trimesters at any point of time, and there would be about 6-7 women in the third trimester. During the course of observations, another 6-7 women from the second trimester will become eligible for the study. Thus, it was estimated that about 14 antenatal women in the third trimester could be included per village. For a sample of 140 antenatal women, we decided to survey 10 villages. The villages were selected purposively from two sub-centers considering similar socioeconomic status, feasibility of visiting these villages in all seasons and availability of health workers.

A total of 141 pregnant women in the third trimester were identified and enrolled in the selected villages. From the antenatal registers available with female workers (anganwari workers) at a village-based Integrated Child Development Services (ICDS), the names of pregnant women were listed out. The key informants in the villages, such as traditional birth attendants, female sarpanches, members of women organizations, etc., were approached to identify additional antenatal mothers. At first visit, the investigator explained the purpose of the study to the pregnant women and took their informed consent. Antenatal mothers were interviewed at their houses using a structured pre-tested questionnaire. It contained questions to record maternal age, education, occupation of the mother and the father, other socioeconomic factors, antenatal care, physical activity during pregnancy and dietary intake in the last 24 h.

The investigator, who was a post-graduate student in community medicine at the time of data collection, had received a weeklong training on food measurements and dietary calculations from the Dietetics Department of PGIMER, Chandigarh. The dietary assessment of pregnant women was done once in every month of the third trimester. The dietary intake from waking up in the morning up to going to bed at night was inquired. Standard containers and weights were used to measure the quantity of intake of cooked food. At each visit, standard dietary advice was given to these antenatal women. Energy intake in terms of kilocalories per day and protein intake in terms of grams per day was calculated at the time of analysis. The nutritional value of food items was as per Indian Council of Medical Research (ICMR) dietary tables.⁷

All the registered pregnant women were followed up at least three times during the third trimester of pregnancy and at delivery. At the time of registration, the weight of pregnant women was recorded by using bathroom-type weighing balance to the nearest 100 g and the height was recorded with the help of anthropometre nearest to the 0.5 cm. A follow-up measurement was done 1 month before delivery. Physical activities in these pregnant women were classified as mild (sedentary), moderate and heavy (hard) as per standard guidelines.⁸ Non-stretchable measuring tape was used to record the height of the mother. Anganwari workers and traditional birth attendants informed the investigator about the delivery by telephone. The investigator took birth weights within 48 h of birth.

Data analysis

Descriptive statistics was used for summarization of data. The outcome variable was low birth weight (weight less than 2500 g). The association among independent variables such as socio-demographic characteristics, economic status, dietary intake, energy expenditure and antenatal care was determined by using Chi-square test for qualitative variables. Quantitative variables were analyzed by Student’s t-test. Multivariate analysis (binary logistic regression analysis) was done to find the ‘independent’ relationship of variables, which were found significant in the bivariate analysis with LBW.

Results

A total of 142 babies were born during the 1 year in the study villages. One lady delivered twins and there were two stillbirths. Weights of the stillborns could not be recorded. The weights of 140 newborn babies could be measured and included in the study.

It was a Hindu (90%) dominated community with 47% backward caste and 24% scheduled caste (23.6%) families. Seventy-nine percent of women lived in joint families. The average family size was 6.3. Eighty-one percent of women were housewives, and 50% of their spouses were labourers. Only 23.6% had a family per capita income of Rs. 1000 or more. Others had a family income much less than that. Twenty-five percent of women were illiterate.

The mean maternal age was 23 years with 22.1% in the age group of 16-20 years and 59.4% in the age group of 21-25 years. Sixty-one percent of newborn babies were male, 10% were pre-term and 9.3% were postterm (over 42 weeks of gestational age). Seventy-four percent of babies were appropriate for gestational age (AGA at ±1 SD), 11% of babies were large for gestational age (LGA) and 17.6% were short for gestational age (SGA) as per the Intra-Uterine Growth Curves Chart of Neonatology Department, PGIMER, Chandigarh.

The mean birth weight was 2786.42 ± 426.3 g. There were 5.2% newborns with birth weight less than or equal to 2000 g, and 28% were above 3000 g. The mean birth weight among term babies was 2840 ± 402 g. It was higher than that of pre-term (2378 ± 417 g) and post-term babies (2753 ± 427 g) (P = 0.001). The prevalence of babies weighing less than 2500 g was 24.3% (95% CI 17.4-32.2%). The prevalence was lower in term babies (18.6%) as compared to pre-term (64.3%) and post-term babies (30.8%).

Calorie intakes were up to 1400 kcal in 12.1% of women, 1401-1,600 kcal in 26.6%, 1601-1800 kcal in 29.3%, 1801-2000 kcal in 24.3% and more than 2000 kcal in 7.7%. The mean caloric intake during three dietary assessments was 1695 ± 182.8 kcal. Maximum (47.2%) number of women received protein in the range of 40-50 g. Only 7% of women received less than 40 g of protein, and 3-6% of women consumed protein greater than 70 g. The mean protein intake during three dietary assessments was 50.8 ± 9.27 g [Table 1].

Significantly higher prevalence (82.7%) of LBW babies was observed in pregnant women with mean caloric intake of less than 1500 kcal, as compared to 12.8% among those consuming at least 1500 kcal/day during the last trimester of pregnancy (P < 0.001). The prevalence of LBW babies among pregnant women with mean protein intake of less than 40 g was significantly higher (60%) as compared to 30.6% among those who were consuming at least 40-50 g proteins/day during the last trimester of pregnancy (P < 0.001) [Table 2]. The mean calorie intake and mean protein intake were not statistically significant in three different gestation groups, i.e. pre-term, term and post-term (P = 0.35 and P = 0.53, respectively).

Bivariate analysis showed a significant association between the birth weight of baby and the maternal age, education, per capita income of family, time of antenatal registration, number of antenatal visits, physical work during pregnancy, height and weight in pregnancy. A significant association between the caloric intake and protein intake with the birth weight of babies [Table 3] was also observed.

Multivariate analysis was carried out between the maternal factors that were significant in bivariate analysis and birth weight of babies. In the binary logistic regression analysis, birth weight was dependent or outcome variable, and significant maternal quantitative factors were independent or predictor variables. In this analysis, birth weight of babies was classified into two categories, i.e. LBW and normal birth weight (NBW). Maternal factors were categorized into three groups (tertiles) by using the categorization method in SPSS analysis package [Table 4].

The relationship between the birth weight of babies and various factors that may inß uence the dietary intake of mothers were observed in the four-stepped models. Four factors, namely caloric intake, protein intake, per capita income and mother’s age, were included in all the four models. In model I, the association of caste, gravidity, maternal education and physical work was tested for significance. It was observed that there was a significant association between the caloric intake, protein intake and physical work of mother with LBW (P-values of 0.02, 0.05 and 0.04, respectively).

In model II, factors like caste and maternal education were dropped since these were found to be non-significant in the first model. Other factors like gestation period at antenatal registration, consumption of non-vegetarian food and number of antenatal visits were added in this model for analysis and were found to be non-significant. Caloric intake and physical work of mother remained significantly associated with the birth weight of babies after taking into consideration the effects of gestational age at registration and non-vegetarian diet. However, with these factors, the association of protein intake with LBW became non-significant.

Table 1: Energy and protein consumption in the third trimester of pregnancy (24 hours dietary recall)

Energy consumption		Protein intake
In kilocalories	N = 140 (%)	In grams	N = 140 (%)
Up to 1400	12.1	<40	7.1
>1400-1600	26.6	≥40 to 50	47.2
>1600-1800	29.3	>50 to 60	32.1
>1800-2000	24.3	>60 to 70	10.0
>2000	7.7	>70	3.6
Mean caloric intake	1695 ± 15.4	Mean protein intake	46.59 ± 9.27

Table 2: Third trimester 24 h caloric and protein intake and birth weight

	Total number	Birth weight <2500 g
Energy consumption (kcal)
<1500	23	19 (82.7)*
≥1500	117	15 (12.8)
Protein intake (g)
<40	10	6 (60.0)*
40-50	62	19 (30.6)
>50	68	9 (13.2)

*P < 0.001, Figures in parentheses are in percentage

Table 3: Risk factors of low birth weight

Risk factors	Mean (±SD)		P-value
	Low birth weight	Normal birth weight
Maternal age (years)	24.441 (±4.19)	22.981 (±3.14)	0.03
Paternal age (years)	26.618 (± 4.34)	25.821 (±3.320)	0.04
Maternal education (years)	4.599 (±4.27)	6.764 (±4.12)	0.01
Paternal education (years)	7.822 (±4.32)	9.064 (±3.48)	0.1
Economic status (per capita income in a month)	674.85 (±31.4)	762.22 (±381.98)	0.02
Time of antenatal registration (months)	4.599 (±1.74)	3.726 (±1.54)	0.01
Average number of antenatal visits	2.471 (±1.23)	3.330 (±1.75)	<0.01
Average gestational age (weeks)	38.625 (±2.55)	39.57 (±2.00)	0.01
Gravida	2.647 (±1.43)	2.104 (±1.22)	0.03
Parity	1.559 (±1.39)	1.019 (±1.07)	0.03
Mean duration of rest per day (h)	7.656 (±1.60)	8.407 (±1.37)	0.02
Mean physical activity per day (h)	7.412 (±3.18)	4.771 (±1.92)	<0.01
Mean weight of mother during early third trimester (kg)	46.232 (±4.81)	51.92 (±7.50)	<0.01
Mean height of mother (cm)	151.08 (±6.54)	154.749 (±5.47)	<0.01
Mean caloric intake (kcal)	1572.0 (±222.34)	1735.0 (±149.09)	<0.01
Mean protein intake (g)	46.559 (±9.45)	52.17 (±8.831)	<0.01
Biofuel exposure (h)	2.118 (±1.14)	1.806 (±1.33)	0.15

Table 4: Binary logistic regression analysis of possible factors associated with nutritional intake and low birth weight

Variables in the equation	Regression coefficient (β)	Standard error (SE)	P-value
Model I
Caloric intake	0.711	0.311	0.02*
Protein intake	0.609	0.311	0.05*
Per capita income	0.316	0.282	0.2
Caste	–0.194	0.153	0.2
Gravidity	0.03	0.299	0.9
Mother’s age	–0.394	0.318	0.2
Maternal education	0.129	0.324	0.6
Physical work	–0.729	0.357	0.04*
Model II
Caloric intake	0.730	0.313	0.02*
Protein intake	0.556	0.314	0.07
Per capita income	0.326	0.283	0.2
Gravidity	0.014	0.308	0.9
Mother’s age	–0.365	0.325	0.2
Physical work	–0.737	0.362	0.04*
Gestation at registration	–0.241	0.283	0.3
Non-vegetarian diet	0.280	0.478	0.5
Antenatal visits	0.503	0.331	1.6
Model III
Caloric intake	0.813	0.33	0.01*
Protein intake	0.584	0.318	0.06
Per capita income	0.132	0.304	0.6
Caste	–0.215	0.161	0.1
Mother’s age	–0.424	0.336	0.2
Physical work	–0.725	0.372	0.05*
Gestation at registration	–0.248	0.29	0.3
Non-vegetarian diet	0.258	0.484	0.5
Mothers height	0.658	0.311	0.03*
Model IV
Caloric intake	0.907	0.340	0.008*
Protein intake	0.448	0.320	0.1
Per capita income	0.176	0.314	0.5
Gravidity	–0.036	0.314	0.9
Mother’s age	0.463	0.340	0.1
Physical work	–0.610	0.375	0.1
Gestation at registration	–0.166	0.306	0.5
Non-vegetarian diet	0.220	0.498	0.6
Mothers height	0.496	0.319	0.1
Mothers weight	0.729	0.329	0.02*

*Significant P-values in various models of mother

In model III, gravidity and number of antenatal visits were dropped and mother’s height was included in the analysis. Height was found to be significantly associated with LBW in addition to caloric intake and physical work. In model IV, mother’s weight was also included in the analysis. With this, the association of height became non-significant, whereas the association of caloric intake (P < 0.01) and weight of the mother (P = 0.02) remains significant and independently associated with LBW.

Discussion

LBW prevalence in this rural population was 24.3%. It was lower than the prevalence (29%) in overall rural Haryana but higher than the national LBW prevalence of 23%.² As compared to the surveys conducted in the same districts in 1980s(3) and 1990s,⁴ the prevalence of LBW has shown a decline of about 1% over the last two decades.

Pregnant women involved in moderate work should consume 2500 kcal/day and 65 g proteins/day.⁶ Very few pregnant women in our study consumed more than 2500 kcal/day during pregnancy. In our study, 82.7% of LBW babies were born to mothers who consumed less than 1500 kcal/day. For mothers who consumed 1500 kcal and above per day, LBW prevalence was only 12.8%. The mean caloric intake of women who delivered LBW babies was significantly lower than those who delivered NBW babies. Other community-based studies from India have shown similar findings.^9-11

Some studies have shown that there is no association between birth size and caloric and protein intake during pregnancy. In a study with rural Indian women who were short and underweight, energy and protein intakes were low at 18 and 28 weeks of gestation. There was a lack of association between size at birth and maternal energy and protein intake.¹²

Maternal dietary composition has an effect on fetal growth.^13,14 However, what type of diet to take at different phases of pregnancy is relatively complex. Energy demand of foetus is higher in the later half of pregnancy. There is some evidence to suggest that this demand is met primarily from the fat deposits that occur in the first half or pre-pregnancy period. It also has a relationship with dietary proteins, especially dairy products in early pregnancy. Thus, the association of low energy intake during the last trimester and lower maternal weight with LBW in our study may indicate that these women had lower fat deposits and still had lower dietary intakes during the first half of pregnancy. However, further studies are required to establish this.

Some nutritional intervention trials have shown improvement in birth weight with caloric supplementation.¹⁵ When mothers in a community were provided with additional 200 kcal and 25 g proteins/day during the last 6-8 weeks of pregnancy, birth weight was more as compared to the non-supplementation group during pregnancy.¹⁶ These evidences suggest that nutritional interventions should be targeted to meet caloric deficits during the third trimester.

To conclude, a marginal decline of 1% in LBW has occurred in this area over the last two decades. Caloric deficit in the third trimester and low maternal weight is associated with higher LBW. The association of this deficit with maternal fat deposition and dietary intake in earlier pregnancy needs to be further established.

The current programmes should focus not only on higher energy intake during third trimester but also on increasing the maternal weight by the time the mothers reach their third trimester through interventions either in pre-pregnant state or in early pregnancy.

References

1. Chaudari S, Kulkarni S, Pandit A, Desmukh S. Mortality and morbidity in high risk infants during six year follow up. Indian Pediatr 2000;37:1314-20.
2. National Family Health Survey (NFHS-II) 1998-99. International Institute for Population Sciences (IIPS) and Operational Research Center (ORC): Mumbai; 2000. 299.
3. Walia I. Neonatal weaning practices (rural community based study) thesis submitted for Ph.D to PGIMER, Chandigarh; 1983.
4. Aggarwal AK, Kumar R. Low birth weight prevalence and antenatal care practices in a rural area of Haryana. Indian Pediatr 1998;35:1031.
5. Kramer MS. Determinants of low birth weight: Methodological assessment and Meta analysis. Bull World Health Organ 1987;65:663-737.
6. World Health Organisation. Technical Report Series No. 522, 1972.
7. National Institute of Nutrition (NIN). Dietary guidelines for Indians. Reprinted by Department of Family Welfare, Ministry of Health and Family Welfare, Govt. of India, Hyderabad, India.
8. Indian Council of Medical Research (ICMR). Recommended dietary intakes for Indians, New Delhi; 1990.
9. Anand K, Garg BS. A study of factors affecting low birth weight. Indian J Community Med 2000;25:57-62.
10. Radhakrishnan T, Thanakappan KK, Vasan RS, Sarma PS. Socioeconomic and demographic factors associated with birth weight: A community based study in Kerala. Indian Pediatr 2000;37:872-6.
11. Murphy JF, Dauncey M, Newcombe R, Garcia J, Elbourne D. Employment in pregnancy, prevention, maternal characteristics and perinateal outcome. Lancet 1984;1: 1163-6.
12. Rao S, Yajnik CS, Kanade A, Fall CH, Margetts BM, Jackson AA, et al. Intake of micronutrient-rich foods in rural Indian mothers is associated with the size of their babies at birth: Pune Maternal Nutrition Study. J Nutr 2001;131:1217-24
13. Moore VM, Davies MJ, Willson KJ, Worsley A, Robinson JS. Dietary composition of pregnant women is related to size of the baby at birth. J Nutr 2004;134:1820-6.
14. Godfrey K, Robinson S, Barker DJ, Barker D, Osmond C, Cox V. Maternal nutrition in early and late pregnancy in relation to placental and fetal growth. BMJ 1996;312: 410-4.
15. Hediger ML, Scholl TO, Ances IG, Beilsky DH, Salmon RW. Rate and amount of weight gain during adolescent pregnancy: Associations with maternal weight for height and birth weight. Am J Clin Nutr 1990;52:793-9.
16. Singh M. Disorders of weight and gestation. In: Singh M. Care of the newborn, 4th ed. Sagar Publications: New Delhi; 1991. p. 28-36.

School of Public Health, Department of Community Medicine, Post Graduate Institute of Medical Education and Research (PGIMER),
Chandigarh – 160 012, India

Correspondence to:
Dr. Arun Kumar Aggarwal, School of Public Health, Department of Community Medicine, 5th Floor, Research ‘B’ Block, Post Graduate Institution of Medical Education and Research (PGIMER), Chandigarh – 160 012, India.
E-mail: aggak63(at)rediffmail.com
Received: 19.09.06
Accepted: 17.11.07