AJCN Tufts Nutrition Symposium, Boston & Online Sept 2009
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.
Agricola
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.
American Journal of Clinical Nutrition, Vol. 70, No. 2, 269-276, August 1999
© 1999 American Society for Clinical Nutrition

Original Research Communications

Energy intake and expenditure of free-living, pregnant Colombian women in an urban setting1,2,3

Darna L Dufour, Julio C Reina and GB Spurr

1 From the Department of Anthropology, University of Colorado, Boulder; the Departments of Physiological Sciences and Pediatrics, Universidad del Valle, Cali, Colombia; and the Department of Physiology, Medical College of Wisconsin and the Research Service, Zablocki Veterans Administration Center, Milwaukee.

2 Supported by NIH grant 5-R22-DK39734. Support for data analysis and manuscript preparation was provided by a CRCW Faculty Fellowship to DLD from the University of Colorado.

3 Reprints not available. Address correspondence to DL Dufour, Department of Anthropology, University of Colorado, Boulder, CO 80309-0233. E-mail: dufourd{at}spot.colorado.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: This study examined the components of energy balance in poor, free-living pregnant women living in an urban setting of a developing country.

Objectives: We tested the following hypotheses: 1) energy intake increases in pregnancy and is greater than when nonpregnant and nonlactating (NPNL), 2) basal metabolic rate (BMR) increases in pregnancy and the increase is positively correlated with prepregnancy fatness, and 3) energy expenditure in activity decreases in pregnancy and is lower than in NPNL women.

Design: Pregnant women were studied at 14.8 ± 3.4 (n = 40), 25.0 ± 3.2 (n = 54), and 34.9 ± 2.4 (n = 43) wk gestation, and NPNL women at baseline (n = 114) and at 3 (n = 103) and 6 (n = 93) mo. Energy intake was measured by using estimated diet records and energy expenditure by using the flex heart rate method. Time allocation in physical activity was assessed by observation.

Results: In pregnant women, body weight, BMR, and energy intake increased but total daily energy expenditure (TDEE) did not change significantly. There were no significant changes in time allocation to selected activities except for lying down. In comparison with NPNL control subjects, women in late pregnancy had higher energy intakes and BMRs. Values for TDEE were not significantly different, but pregnant women expended less energy in activity and allocated more time to 2 energy-saving activities and less time to 2 energy-demanding activities.

Conclusion: A decrease in energy expenditure in activity and changes in time allocation are important ways in which pregnant women meet the energy demands of pregnancy.

Key Words: Energy intake • energy expenditure • diet • nutrition • developing country • pregnancy • urban setting • heart rate • activity • time allocation • poverty • basal metabolic rate


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Pregnant women need extra energy to cover the energy demands of the fetus and associated tissues and their maintenance (1). In all but the most extreme circumstances, pregnant women deposit adipose tissue reserves (2), a function that also requires additional energy. Several studies have shown that energy intake does not increase sufficiently to meet the theoretic needs (3). This discrepancy between the energy cost of pregnancy and measured dietary intake has led to the conclusion that either the estimates of energy intake are flawed or pregnant women are able to lower their total energy expenditure and, hence, their energy needs.

Basal metabolism is the best-studied component of energy expenditure in pregnancy. Longitudinal studies of free-living women in Europe, Thailand, Philippines, and Gambia (4) have shown increases in basal metabolism and together have provided evidence of a marked plasticity in basal metabolism in pregnancy (5). This finding has led to the suggestion that women in developing countries who begin pregnancy with less body fat will have a more energy-sparing metabolic response to pregnancy than will women in more affluent countries who begin pregnancy with greater fat stores (5).

Another major component of total energy expenditure, energy expenditure in activity, has proven difficult to measure in free-living women. Earlier studies in Europe, Thailand, and Philippines used the activity diary method to estimate energy expenditure in activity but the method was found to lack precision (4). More recent studies in Europe (6, 7) used the doubly labeled water (DLW) technique to assess total energy expenditure and energy expenditure in activity as the difference between total energy expenditure and basal or resting metabolism. Except for studies of marginally nourished women in Gambia (8, 9), we are unaware of comparable studies of total energy expenditure and energy expenditure in activity in free-living women in developing countries.

The purpose of this study was to examine the components of energy balance in free-living pregnant women living in an urban setting in a developing country and to compare them with those in nonpregnant, nonlactating (NPNL) women. We tested the following hypotheses: 1) energy intake is greater in late pregnancy and is greater in pregnant than in NPNL women, 2) basal metabolism increases in pregnancy and the increase is positively correlated with prepregnancy fatness, and 3) energy expenditure in activity is lower in late pregnancy and lower in pregnant than in NPNL women. This is the only study to present data on the energy metabolism of pregnant women in an urban setting in a developing country and to examine data on both energy metabolism and time allocation in pregnant women.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects and setting
The subjects were pregnant and NPNL women aged 19–43 y, all of whom were volunteers recruited via word of mouth and were apparently healthy as judged by a routine medical examination performed by a local physician. The research protocol was approved by the Human Research Review Committee of the Medical College of Wisconsin and the Research Committee of the Universidad del Valle. All the women were part of a larger study of energy metabolism and nutrition of poor urban women.

The women lived in 2 poor neighborhoods (barrios) on the periphery of Cali, a city of {approx}2.2 million people. Both neighborhoods were originally settled in the mid-1970s and were representative of the unplanned and unregulated urbanization that accompanied the massive migration from rural to urban areas that occurred in Colombia at that time. The residential units in both neighborhoods were small houses. By 1994, one of the neighborhoods was composed of mostly brick houses, and all were connected to municipal electricity, water, and sewerage. The second neighborhood had developed more slowly and in 1994 many of the houses were still under construction; some were temporary structures made of recycled materials. Most of the houses had electricity and water through connections the residents had engineered themselves. Residents in some areas had also installed sewer pipes. Both neighborhoods had a government-supported health post and an elementary school. Neither had paved streets, adequate street lighting, or telephone service, but both had solid waste disposal provided by the city. However, in a Pearson chi-square analysis of most socioeconomic variables (eg, house ownership and construction and kitchen utensils), one of the neighborhoods consistently scored lower on the socioeconomic scale than the other. The lower-scoring neighborhood also had a higher percentage of women of Afro-Colombian descent.

Study design
The pregnant women were studied in 3 measurement rounds, which corresponded to 14.8 ± 3.4 (n = 40; range = 6.7–18.8), 25.0 ± 3.2 (n = 54; range = 19.0–29.8), and 34.9 ± 2.4 (n = 43; range = 30.4–39.4) wk gestation. Pregnancy weeks were calculated from the date of the last menstrual period obtained by recall at the baseline interview. The NPNL women were also studied in 3 measurement rounds at baseline (n = 114) and {approx}3 (n = 103) and 6 (n = 93) mo. All measurements in each round were completed during a 1-wk (5 weekdays) period.

Anthropometric measurements, interviews, and measurements of resting metabolic rate (RMR) and metabolic rate in activity (treadmill walking) were done in the laboratory on 1 day. Measurements of basal metabolic rate (BMR) were done in duplicate on 2 other consecutive days in the early mornings in the subjects' homes. After the measurement of BMR, each subject was free to carry on with her daily activities as she normally did, but wore a heart rate recorder until she retired for the night. In addition, her physical activities and food intake were recorded by a trained observer from the neighborhood. This design resulted in data for 2 consecutive days of food intake, energy expenditure, and time allocation for each woman in each of the 3 measurement rounds.

Anthropometry, blood hemoglobin, and sociodemographic characteristics
Anthropometric measurements were taken by a trained technician following standard techniques described by Lohman et al (10). Body weight was measured to ±25 g with a Homs Beam Balance (Hom Corp, Belmont, CA) with the subjects lightly clothed and without shoes. Height was obtained with a wall stadiometer. Skinfold thicknesses were measured in triplicate by using a Lange skinfold caliper (Cambridge Scientific Instruments, Cambridge, MD), and midupper arm, thigh, and calf circumferences were measured with a flexible steel tape. Body mass index (BMI) was calculated as weight/height2. Fat mass was estimated with an empirical equation derived from women in this population (11). Estimations were done for round 1 only because the equation was derived from NPNL women and has yet to be validated for use in pregnant women. The changes in tissue composition that occur in pregnancy make it difficult to measure changes in maternal body composition with any of the techniques now available for use in field studies (12). Blood hemoglobin was determined with fingertip blood samples by the cyanmethoglobin method. Birth weights of the infants born to women in the study were obtained from hospital records. Sociodemographic information on age, parity, educational level, marital status, and employment was obtained by structured interview, and information about living conditions was obtained by observation.

Dietary intake
Dietary intake data were collected by using estimated food records kept by trained observers. The observers were young women (aged 18–25 y) who lived in the same neighborhoods as the subjects and were trained to unobtrusively record both physical activity and dietary intake. They worked in pairs and alternated 4-h shifts scheduled to cover the awake portion of subjects' days ({approx}16 h). The accuracy of the observers in estimating serving sizes was assessed periodically.

In recording food intake, observers estimated the serving size of food items as they were served to or selected by the subject. For foods served by the piece, such as pieces of meat or fruit, the observers recorded a description of the food and its relative size (small, medium, or large). For foods served with use of spoons or cups, observers recorded the type of spoon or cup used and the degree of fullness. Descriptions of food item sizes were later converted to gram weights by using a food-portion-size data table developed specifically for the study. In addition, volumetric measurements of serving utensils and drinking vessels were completed in each subject's home before observations of the food intake for that subject began, and these measurements were used as appropriate. Observers were also trained to obtain serving size weights of all new and unusual foods as well as common foods served with unusual utensils and to add this data to the portion size table as the study progressed.

The energy content of foods was obtained from published food-composition tables (13–15) and proximate analyses of samples of the most commonly consumed foods. The latter were completed by the Instituto Nacional de Nutrición in Bogotá. Recipes (in household units) were collected for composite dishes and beverages. For the more common dishes and beverages, weighed recipes were collected, the food was prepared in the laboratory to determine the final cooked weight of the recipe, and the nutrient composition was calculated from published tables or proximate analyses. For packaged snack foods, nutrient composition was obtained from information on the package. Energy and nutrient intakes were calculated by using custom software programs written by GBS.

Basal metabolic rate and total daily energy expenditure
BMR was measured via indirect calorimetry. The method used was described in detail by Spurr et al (16). Total daily energy expenditure (TDEE) was measured by using the flex heart rate method. This method involves the establishment of the heart rate–oxygen consumption relation for each subject at rest and in activity in each measurement round, the determination of BMR via indirect calorimetery, and the measurement of heart rate at 1-min intervals during the awake portion of the day. Spurr et al (17) described the technique and its use in Cali women (16, 18). RMR was obtained by repeating measurements until oxygen consumption leveled off at minimum rates (differences <=30 mL/min). RMR was calculated as the mean of the values obtained in 3 resting positions (lying down, sitting, and standing quietly). TDEE was calculated as the sum of energy expenditure while the subject was wearing the heart rate monitor, BMR during the hours of sleep (8.5 ± 0.9 h), and RMR for the time (average: 67 min) not accounted for by heart rate monitoring or sleep. TDEE - BMR was defined as the energy expenditure in activity (EEAct). Physical activity level (PAL) was calculated as TDEE/BMR.

The flex heart rate method has been validated by comparison with both whole-body indirect calorimetry (17, 19) and the DLW method in adults (20) and children (21). The DLW method is considered the gold standard in measurements of energy expenditure of free-living subjects. Heini et al (8) found good overall agreement between the flex heart rate and the DLW method in pregnant women.

Muscular efficiency
Given that an increase in the efficiency of muscular work would allow for physical activity to occur at a lower energy cost, we measured muscular efficiency in a subsample of pregnant (n = 9) and NPNL (n = 18) women. Measurements were made on a cycle ergometer as described by Spurr et al (22). The percentage muscular efficiency was calculated as delta efficiency, which is considered the best estimate of muscular efficiency of an individual (23). It is defined as the change in work accomplished/change in energy expended x 100.

Time allocation in physical activity
A record of the activities engaged in by subjects during normal waking hours (14.3 ± 0.7 h) was kept by the same trained local observers who recorded dietary intake. The observers recorded activities using the activity diary technique (24) and a 3-level code. The first level of the code specified body position as lying, sitting, standing, milling about, or walking (slow, moderate, or fast). "Milling about" was used for indoor activities in which women spent time standing as well as walking a few paces across a room. The second and third levels of the code specified culturally defined activities such as washing dishes and ironing. In the analysis presented here we considered the first-level codes only.

Data analysis
In this analysis, we used the mean energy intake for 2 d in each round and the mean TDEE for the same 2 d in each round. For subjects who did not have 2 d of TDEE measurements the value for a single day was used. This occurred in {approx}50% of the subjects in any given round because of the technical difficulties of recording heart rates. In any event, there were no statistically significant differences between the 2 successive days of TDEE measurements in any of the groups of subjects (16). Descriptive statistics are presented as means and SDs unless indicated otherwise. Mean differences between pregnant and NPNL women in anthropometric characteristics, energy intake, BMR, TDEE, EEAct, PAL, and time allocation were each evaluated by two-way analysis of covariance (ANCOVA). The covariates were 3 variables assumed to be related to components of energy balance: 1) employment status; 2) neighborhood of residence, which correlated roughly with economic resources; and 3) age. Pearson chi-square analysis was used to verify between-group differences in the first and second covariates, and a t test was used for the third covariate. Significant between-group and time-by-group interactions in the ANCOVA were followed up by using the Bonferroni test. Bartlett's test for equality of variances was used to test the null hypothesis that variances were equal. When a statistically significant Bartlett's test resulted, the variables were converted to logarithms and the ANCOVA was repeated. All of the above statistical analyses and the power calculations were completed by using STATA statistical software, release 5.0 (Stata Corporation, College Station, TX).

For a subsample of 20 pregnant women, each of whom completed all measurements in each of the 3 rounds, changes in BMR between rounds 1 and 3 were calculated and correlated with measurements of body fatness in round 1. For a second subsample of 9 pregnant women and 18 NPNL control subjects, each of whom completed measurements of muscular efficiency in each of the 3 rounds, mean differences between groups were evaluated by repeated-measures analysis of variance (ANOVA) using a general linear model procedure. Wilks' lambda was used as the criterion of significance and Levene's statistic for equality of variances was used to test the null hypothesis that variances were equal. Pearson correlation coefficients and repeated-measures ANOVA were calculated by using SPSS 7.5 (SPSS Inc, Chicago). Significance was set at P = 0.05 for all statistical tests. Data were plotted by using SIGMA PLOT 4.0 (SPSS Inc).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Sociodemographic characteristics
The pregnant women were somewhat younger than the NPNL women (25.2 ± 5.5 and 28.4 ± 6.1 y, respectively; P = 0.005) and had fewer children (1.0 ± 0.9 compared with 1.6 ± 1.1; P < 0.001). Most (82%) of the pregnant women lived with a spouse or male partner. The mean number of years of formal education was similar in both groups: 6.4 y for pregnant and 6.5 y for NPNL women. Sixty-eight percent of the pregnant women either owned or rented the house in which they were living, and the remainder lived in rented rooms, with relatives or friends without paying rent, or in temporary structures built on someone else's land. These residence characteristics are reflective of relative economic disadvantage in a society where home ownership is the goal. Also indicative of their poverty, 68% of the women lived in houses in which the average number of people per bedroom was >=3. Living conditions of the NPNL women were similar.

More than half of the women in both groups were employed in income-generating activities in each measurement round. These income-generating activities were the same in both groups and included domestic service, street vending, child care, clothes washing, and sales of one type or another.

Anthropometry and hemoglobin status
The mean height of the pregnant women (round 1) was 156.7 ± 6.9 cm and was not significantly different from that of the NPNL women. The mean height was about the 15th percentile of the US National Center for Health Statistics reference population (25). Of the pregnant women, 18% were <150 cm tall, the cutoff for high risk of complications in pregnancy (26). Body fat composition was 35.8 ± 4.5% of body weight in pregnant women in round 1 and not significantly different from that of NPNL women. Other anthropometric characteristics of the pregnant and NPNL women in the 3 measurement rounds are presented in Table 1Go. The mean body weight of pregnant women was significantly greater in round 3 than in round 1 (P < 0.001) as was BMI (P < 0.001). The only skinfold-thickness measurement to show a significant difference between rounds 1 and 3 was the suprailiac skinfold (P = 0.001). None of the circumferences measured were significantly different between the 3 rounds. Differences between pregnant and NPNL women were significant for body weight, BMI, and suprailiac skinfold thickness in rounds 2 and 3.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Anthropometric characteristics and blood hemoglobin of pregnant and nonpregnant, nonlactating (NPNL) women1
 
In pregnant women, mean blood hemoglobin values were >112 g/L in all 3 rounds. The value 112 g/L is the WHO (27) cutoff for anemia in pregnancy adjusted +2 g/L for the altitude of Cali (976 m), as suggested by the Centers for Disease Control and Prevention guidelines (28). The percentage of pregnant women with hemoglobin values <112 g/L was 32%, 46%, and 33% in rounds 1, 2, and 3, respectively. The mean blood hemoglobin concentration was significantly lower in pregnant than in NPNL women in all 3 rounds (P < 0.001) and the pregnant women showed a significantly higher incidence of anemia in all 3 rounds (Pearson chi-square, P < 0.001).

In each of the 3 measurement rounds, 40–50% of the pregnant women were observed to take an iron or multivitamin supplement. In the NPNL women, 12–20% had hemoglobin concentrations <122 g/L, the altitude-adjusted cutoff for nonpregnant women (27, 28) in the 3 measurement rounds. The mean birth weight of live-born infants (n = 37) was 3170 ± 515 g. Of these, 4 infants weighed <2500 g and would be classified as having low birth weight.

Energy intake
Means and SDs for energy intakes and expenditures are presented in Table 2Go. In pregnant women, energy intake in round 3 was 1.36-MJ higher than in round 1 (P = 0.040). Absolute values for energy intake were higher in pregnant than in NPNL women in rounds 2 and 3, and differences were significant in round 3 (P = 0.003).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Energy intake and components of energy expenditure of pregnant and nonpregnant, nonlactating (NPNL) women1
 
Energy expenditure
Mean values for BMR in pregnant women were 11% higher in round 3 than in round 1 (P = 0.004) (Table 2Go). Mean BMR in pregnant women was significantly higher than in NPNL women in rounds 2 (P = 0.003) and 3 (P < 0.001). In a subsample of 20 pregnant women with measurements of BMR in each round, the change in BMR (round 3 - round 1) was not significantly correlated with percentage body fat in round 1 nor change in body weight between rounds 1 and 3.

In pregnant women, mean values for TDEE did not change significantly over the course of the 3 measurement rounds. Mean values for TDEE were not significantly different in pregnant and NPNL women. In pregnant women, the mean values for EEAct and PAL, usually considered measures of energy expenditure in physical activity, tended to decrease over time, but differences were not significant. Mean PAL values were lower in pregnant than in NPNL women in rounds 2 (P = 0.001) and 3 (P = 0.01).

Muscular efficiency
Percentage muscular efficiency (expressed as delta efficiency) in a sample of pregnant women (n = 9) was 28.7 ± 3.6%, 29.4 ± 2.8%, and 28.1 ± 3.2% and in NPNL (n = 18) women was 26.2 ± 2.3%, 27.1 ± 2.0%, and 28.5 ± 3.4% (x ± SE) in rounds 1, 2, and 3, respectively. There were no significant differences between the mean values for the 3 rounds of measurements, nor were there any significant differences between the pregnant and NPNL women.

Time allocation in physical activity
Time allocation in physical activity is shown in Figure 1Go. The only significant difference in time allocation in pregnant women in the 3 measurement rounds was an increase in time spent lying down between rounds 1 and 3 (P = 0.02). There were, however, significant differences in time allocation between the pregnant and NPNL women in 5 of the 6 activities measured. Pregnant women spent significantly more time lying down (all rounds), more time sitting (round 2), less time standing (all rounds), and less time milling about (all rounds) than did NPNL women. Pregnant women also spent less time walking at a moderate-to-fast pace (rounds 2 and 3) than did the NPNL women.



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 1. Mean (±SEM) time allocation in activities of pregnant and nonpregnant, nonlactating (NPNL) women. n = 40, 54, and 43 pregnant women and n = 114, 103, and 93 NPNL subjects measured in rounds 1, 2, and 3, respectively. Note that the scale of the lower graphs is different from that of the upper graphs. *Significant difference between pregnant and NPNL, P < 0.05 (post hoc Bonferroni test).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The women in this study were volunteers but we think they are representative of a large segment of the urban poor in Colombia because they share many similarities in living conditions, lifestyles, and occupations with the urban poor described in other studies (30, 31). The mean birth weight of infants born to pregnant women in this study compared favorably with other data from Colombia, and the frequency of low-birth-weight infants was lower than the Colombian average (32).

On the basis of anthropometric characteristics, the nutritional status of the women was normal. Their short stature, however, is indicative of undernutrition during the period of growth. In pregnant women, weight attained by round 3, 63.1 kg, was at about the 50th percentile for Colombian women in the ninth lunar month of pregnancy (32). The higher mean body weight of pregnant women in round 3 was paralleled by greater skinfold thicknesses at the suprailiac site. This skinfold is a typical site of fat deposition in pregnancy (12), and the greater values found in Cali women in round 3 are interpreted as evidence of increased maternal fat stores. The lower blood hemoglobin values of the pregnant women are assumed to be the result of the normal expansion of plasma volume that occurs in pregnancy (33).

Our first hypothesis, that energy intake increases in pregnancy and energy intake is greater in pregnant than in NPNL women, was supported. In pregnant women, energy intake in round 3 was significantly higher than in round 1, and intake in round 3 was higher in pregnant than in NPNL women. For comparison of group means and SDs, with n = 46 pregnant and n = 93 NPNL women, the power was 0.79.

Studies of European and Australian women assumed to be eating to appetite (6, 34, 35) have not shown the expected increase in energy intake in pregnancy. In all these studies the subjects themselves kept the diet records, and in at least one study there is clear evidence of underreporting of intake (6). The Cali women cannot be assumed to be eating to appetite because they were a subsample of a larger group that exhibited considerable evidence of food insecurity (36). Food intake was recorded by observers and we have no indication that intake was underrecorded. Other studies in developing countries where pregnant women may not have been able to eat to appetite have provided contradictory results: rural women in Philippines showed a decrease in energy intake (37), and rural women in Thailand showed a moderate increase in energy intake (38).

Although pregnant women could theoretically cover the cost of pregnancy by reducing TDEE, we found no differences in TDEE among the 3 measurement rounds in pregnant women in this study, nor any difference between pregnant and NPNL women. With a power of 0.80, we would have been able to detect between-group differences of 2.7-, 1.5-, and 2.3-MJ/d in rounds 1, 2, and 3, respectively, but the differences we found were smaller. We expected TDEE to be lower than energy intake, but differences were not significant in rounds 2 or 3. As Goldberg et al (6) pointed out, the real and considerable day-to-day variation in both energy intake and expenditure results in a degree of uncertainty in both measures.

The 2 components of TDEE, BMR and EEAct (TDEE - BMR), showed opposite tendencies over the 3 rounds of measurements. BMR increased significantly, but this increase was counterbalanced by a nonsignificant tendency for EEAct to decrease. The net result was no change in TDEE. The 11% increase in BMR found in this study is less than that reported for rural women in Philippines (37) and Thailand (38), and less than the increases in BMR reported for European women (4, 6). Neither our study nor those in Thailand and Philippines had measurements of prepregnancy BMR and so may have underestimated the total increase in BMR. Furthermore, the relatively late entry of some of the women into our study meant that increases in BMR could have occurred before our first measurement. Late enrollment reflected the difficulties of recruiting pregnant women in a community in which women typically do not seek prenatal care in the first trimester. In Cali women we found no evidence to support the proposal that small increases in BMR in pregnancy would be more likely in women who begin pregnancy with low energy reserves (39).

Although a decrease in EEAct is considered one of the most important mechanisms by which women cover the additional energy costs of pregnancy, the decrease has been difficult to show convincingly because of the inherent difficulties of measuring energy expenditure in spontaneous physical activity in free-living women. In this study, we measured EEAct indirectly as TDEE - BMR, which represents the sum of EEAct and diet-induced thermogenesis (DIT). If we assume that DIT remains essentially unchanged in pregnancy (3, 40), and we know that TDEE did not change, then decreases in EEAct should reflect reductions in energy expenditure in activity. In pregnant Cali women absolute values for EEAct were 18% less than those of NPNL women, a difference that could be of biological significance even though it was not statistically significant.

PAL (or TDEE/BMR), is another measure of energy expenditure in activity. In pregnant Cali women, PAL values dropped from 1.9 in round 1 to 1.7 in round 3, and were significantly lower in both rounds 2 and 3 than in NPNL women. Although a decrease in PAL values suggests a decrease in energy expended in physical activity, PAL values will decrease in pregnancy even if physical activity remains constant because BMR increases. In well-nourished European women PAL values were lower at 37 wk by {approx}0.15 units (3). This is less than the change seen in Cali women, again suggesting that energy expenditure in activity declined.

An increase in muscular efficiency in pregnancy would allow for physical activity to occur at a smaller energy cost and could therefore be energy sparing. However, we found no change in muscular efficiency (measured as delta efficiency) in pregnant women during gestation and no difference between pregnant and NPNL women. These findings agree with those of Poppitt et al (9) for pregnant Gambian women.

Time allocation is a measure of behavioral choices with respect to physical activity. We found that pregnant Cali women spent more time lying down in round 3 than in round 1, but otherwise did not change the time devoted to the activities measured. There were, however, some clear differences between pregnant and NPNL women. Pregnant women spent more time lying down and less time standing and milling around. These differences were evident in round 1 (14.8 ± 3.4 wk gestation), suggesting that changes in activity may occur very early on in pregnancy. Furthermore, pregnant women spent less time walking at a moderate-to-fast pace. This finding agrees with those of 2 other studies that showed reductions in the pace of walking in pregnant women (41, 42).

The differences in time allocation between pregnant and NPNL women in round 1 would theoretically result in a small energy savings for pregnant women. However, in late pregnancy (round 3) this savings would be lost because the increase in BMR means that energy expenditure in all activities increases. In a sense then, the differences in time allocation seen in pregnant women function to hold TDEE relatively constant while BMR increases. Our estimates of energy expenditure in selected activities are approximations and cannot account for differences in the intensity of activity, which could, as Durnin (43) has suggested, lead to significant savings.

Our results from Cali are difficult to compare with those of other studies of time allocation in pregnancy because activities are categorized in different ways in different studies. However, several other studies in developing countries have shown a shift in pregnancy toward activities with lower energy costs (36, 44, 45). Studies from developed countries, on the other hand, have shown conflicting results and not necessarily shifts in the direction expected (4, 46). None of these studies have quantified the reductions in energy expenditure attributable to the change in time allocation, so it is difficult to know how much of a savings in energy might be expected.

In summary, in these pregnant, economically disadvantaged urban women in Colombia who were measured at {approx}15, 25, and 35 wk gestation, energy intake was higher in late pregnancy, but there were no significant differences in TDEE between early and late pregnancy. There were no significant differences in time allocation of major coded activities among the 3 measurement rounds, except for a greater amount of time spent lying down in rounds 2 and 3. In comparison with NPNL control subjects, pregnant women had higher energy intakes and BMRs in rounds 2 and 3, but values for TDEE that were not significantly different. PAL values were lower in pregnant women in rounds 2 and 3 and these women allocated more time to energy-saving activities and less time to more energy-demanding activities. These differences support the suggestion that changes in physical activity are an important way pregnant women meet the energy demands of pregnancy.


   ACKNOWLEDGMENTS
 
We express our gratitude to Zoila de Maldonado, Betty de Orozco, and Jairo Ardila for their excellent technical support and also to Blanca Diaz, Clara Ines Gil Rivas, Mayerling Valencia Castro, Ruth Garcia Ramirez, and Yasmin Villota for their invaluable assistance in data collection. We are also grateful to Secretaría del Deporte y Recreación de Cali for logistical support, to the Instituto Nacional de Nutrición for the proximate analyses of common foods, and to Fundación Arcesio Paz Paz for their support. We thank the women of Agua Blanca for their patience and cooperation.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hytten FE. Weight gain in pregnancy. In: Hytten FE, Chamberlain G, eds. Clinical physiology and obstetrics. Oxford, United Kingdom: Blackwell Scientific Publications, 1991:173–203.
  2. Lawrence M, Coward WA, Lawrence F, Cole TJ, Whitehead RG. Fat gain during pregnancy in rural African women: the effect of season and dietary status. Am J Clin Nutr 1987;45:1442–50.[Abstract/Free Full Text]
  3. Prentice AM, Spaaij CJK, Goldberg GR, et al. Energy requirements of pregnant and lactating women. Eur J Clin Nutr 1996;60(suppl):S82–111.
  4. Durnin JVGA. Energy requirements of pregnancy: Lancet 1987;2:895–6.[Medline]
  5. Prentice AM, Goldberg GR, Poppitt SD. Reproductive stresses in undernourished and well-nourished women. In: Porrini M, Walter P, eds. Nutrition in pregnancy and lactation. Bibl Nutr Dieta. Basel: Karger, 1996;53:1–10.
  6. Goldberg G, Prentice AM, Coward WA, et al. Longitudinal assessment of energy expenditure in pregnancy by the doubly labeled water method. Am J Clin Nutr 1993;57:494–505.[Abstract/Free Full Text]
  7. Forsum E, Kabir N, Sadurskis A, Westerterp K. Total energy expenditure of healthy Swedish women during pregnancy and lactation. Am J Clin Nutr 1992;56:334–42.[Abstract/Free Full Text]
  8. Heini A, Schutz Y, Diaz E, Prentice AM, Whitehead RG, Jéquier E. Free-living energy expenditure measured by two independent techniques in pregnant and nonpregnant Gambian women. Am J Physiol 1991;261:E9–17.[Abstract/Free Full Text]
  9. Poppitt SD, Prentice AM, Jéquier E, Schutz Y, Whitehead RG. Evidence of energy sparing in Gambian women during pregnancy: a longitudinal study using whole-body calorimetry. Am J Clin Nutr 1993;57:353–64.[Abstract/Free Full Text]
  10. Lohman TG, Roche AF, Martorell R, eds. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics, 1988.
  11. Spurr GB, Reina JC, Li SJ, de Orozco B, Dufour DL. Body composition of Colombian women. Am J Clin Nutr 1994;60:279–85.[Abstract/Free Full Text]
  12. Norgan NG. Maternal body composition: methods for measuring short-term changes. J Biosoc Sci 1992;24:367–77.[Medline]
  13. Instituto Nacional de Nutrición. Tabla de Composición de Alimentos Colombianos. (Table of composition of Colombian food.) 3rd ed. Bogota, Colombia: Instituto Nacional de Nutrición, 1967.
  14. Pennington JAT, Church HN. Food values of portions commonly used. 14th ed. New York: Harper & Row Publishers, 1985.
  15. US Department of Agriculture: USDA Nutrient Database for Standard Reference. World Wide Web: http://www.nal.usda.gov/fnic/cgi-bin/nut-search.pl (accessed April 1994).
  16. Spurr GB, Dufour DL, Reina JC. Energy expenditure of urban Colombian women: a comparison of patterns and total daily expenditure by the heart rate and factorial methods. Am J Clin Nutr 1996;63:870–8.[Abstract/Free Full Text]
  17. Spurr GB, Prentice AM, Murgatroyd PR, Goldberg GR, Reina JC, Christman NT. Energy expenditure from minute-by-minute heart-rate recording: comparison with indirect calorimetry. Am J Clin Nutr 1988;48:552–9.[Abstract/Free Full Text]
  18. Spurr GB, Dufour DL, Reina JC, Hoffmann RG, Waslien CI, Staten LK. Variation of the basal metabolic rate and dietary energy intake of Colombian women during 1 y. Am J Clin Nutr 1994;59:20–7.[Abstract/Free Full Text]
  19. Ceesay SM, Prentice AM, Day KC, et al. The use of heart rate monitoring in the estimation of energy expenditure: a validation study using indirect whole-body calorimetry. Br J Nutr 1989;61:175–86.[Medline]
  20. Livingstone MBE, Prentice AM, Coward WA, et al. Simultaneous measurement of free-living energy expenditure by the doubly labeled water method and heart-rate monitoring. Am J Clin Nutr 1990;52:59–65.[Abstract/Free Full Text]
  21. Livingstone MBE, Coward WA, Prentice AM, et al. Daily energy expenditure in free-living children: a comparison of heart-rate monitoring with the doubly labeled water (2H218O) method. Am J Clin Nutr 1992;56:343–52.[Abstract/Free Full Text]
  22. Spurr GB, Dufour DL, Reina JC. Increased muscular efficiency during lactation in Colombian women. Eur J Clin Nutr 1998;52:17–21.[Medline]
  23. Gaesser GA, Brooks GA. Muscular efficiency during steady state exercise: effects of speed and work rate. J Appl Physiol 1975;38:1132–9.[Abstract/Free Full Text]
  24. Durnin JVGA, Passmore R. Energy, work and leisure. London: Heinemann, 1967.
  25. Frisancho AR. Anthropometric standards for the assessment of growth and nutritional status. Ann Arbor, MI: University of Michigan, 1990.
  26. Rey H, Ortiz EI, Fajardo L, Pradilla A. Maternal anthropometry: its predictive value for pregnancy outcome. World Health Organ Bull 1995;73(suppl):70–1.
  27. World Health Organization. Nutritional anaemias. Report of a WHO Scientific Group. World Health Organ Tech Rep Ser 1968;405.
  28. Centers for Disease Control. CDC criteria for anemia in children and childbearing-aged women. MMWR Morb Mortal Wkly Rep 1989;38:400–4.[Medline]
  29. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993;25:71–80.[Medline]
  30. World Bank. Colombia: social programs for the alleviation of poverty. Washington, DC: The World Bank, 1990.
  31. World Bank. Colombia: poverty in Colombia. Washington, DC: The World Bank, 1994.
  32. World Health Organization. Maternal anthropometry and pregnancy outcomes. Bull World Health Organ 1995;73(suppl):1–69.
  33. Pitkin RM. Nutritional influences during pregnancy. Med Clin North Am 1977;61:3–15.[Medline]
  34. Durnin JVGA, McKillop FM, Grant S, Fitzgerald G. Energy requirements of pregnancy in Scotland. Lancet 1987;2:897–900.[Medline]
  35. Truswell AS, Ash S, Allen JR. Energy intake during pregnancy. Lancet 1988;1:49 (letter).
  36. Dufour DL, Staten LK, Reina JC, Spurr GB. Living on the edge: dietary strategies of economically impoverished women in Cali, Colombia. Am J Phys Anthropol 1997;102:5–15.[Medline]
  37. Tuazon MAG, van Raaij JMA, Hautvast JGAJ, Barba CV. Energy requirements of pregnancy in the Philippines. Lancet 1987;2:1129–31.[Medline]
  38. Thongprasert K, Tanphaichitre V, Valyasevi A, Kittigool J, Durnin JVGA. Energy requirements of pregnancy in rural Thailand. Lancet 1987;2:1010–2.[Medline]
  39. Prentice AM, Goldberg GR, Davies HL, Murgatroyd PR, Scott W. Energy sparing adaptations in human pregnancy assessed by whole body calorimetry. Br J Nutr 1989;62:5–22.[Medline]
  40. Singh J, Prentice AM, Diaz E, et al. Energy expenditure of Gambian women during peak agricultural activity measured by the doubly-labelled method. Br J Nutr 1989;62:315–29.[Medline]
  41. Nagy LE, King JC. Energy expenditure of pregnant women at rest or walking self-paced. Am J Clin Nutr 1983;36:369–76.
  42. van Raaij JMA, Schonk CM, Vermaat-Miedema SH, Peek MEM, Hautvast JGAJ. Energy cost of walking at a fixed pace and self-paced before, during and after pregnancy. Am J Clin Nutr 1990;51: 158–61.[Abstract/Free Full Text]
  43. Durnin JVGA. Energy requirements of pregnancy. Diabetes 1991;40(suppl)152–6.
  44. Roberts SB, Paul AA, Cole TJ, Whitehead RG. Seasonal changes in activity, birth weight and lactational performance in rural Gambian women. Trans R Soc Trop Med Hyg 1982;76:668–78.[Medline]
  45. Panter-Brick C. Seasonality of energy expenditure during pregnancy and lactation for rural Nepali women. Am J Clin Nutr 1993;57:620–8.[Abstract/Free Full Text]
  46. van Raaij JMA, Schonk CM, Vermaat-Miedema SH, Peek MEM, Hautvast JGAJ. Energy cost of physical activity throughout pregnancy and the first year postpartum in Dutch women with sedentary lifestyles. Am J Clin Nutr 1990;52:234–9.[Abstract/Free Full Text]
Received for publication June 17, 1998. Accepted for publication February 2, 1999.




This article has been cited by other articles:


Home page
 Int J EpidemiolHome page
S. A Kim, K. M Yount, U. Ramakrishnan, and R. Martorell
The relationship between parity and overweight varies with household wealth and national development
Int. J. Epidemiol., February 1, 2007; 36(1): 93 - 101.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.
Agricola
Right arrow Articles by Dufour, D. L
Right arrow Articles by Spurr, G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS