Paper | Published:

Gestational weight gain and postpartum behaviors associated with weight change from early pregnancy to 1 y postpartum

International Journal of Obesity volume 27, pages 117127 (2003) | Download Citation



OBJECTIVES: (1) To describe the relative importance of gestational weight gain, postpartum exercise, food intake and breastfeeding to weight change from early pregnancy to 1 y postpartum; and (2) to identify subgroups of women at greatest risk for major weight gain surrounding childbearing.

DESIGN: A prospective cohort study of women who registered for obstetrical care in a hospital and primary care clinic system serving a 10 county area of upstate New York.

SUBJECTS: A total of 540 healthy adult women who gave birth to full-term singleton infants.

MEASUREMENTS: Sociodemographic characteristics, exercise, food-related behaviors and breastfeeding were assessed using the medical record and a mailed questionnaire. Body weight was measured at prenatal visits and 1 y postpartum. Weight retained and major weight gain (4.55 kg) at 1 y postpartum were the main outcomes.

ANALYSIS: Linear and logistic regression analyses were conducted.

RESULTS: Women were on average 1.51±5.95 kg heavier at 1 y postpartum than they were in early pregnancy. Nearly 25% of women experienced a major weight gain of 4.55 kg or more at 1 y postpartum. Gestational weight gain, exercise frequency, change in food intake and breastfeeding were each significantly related to postpartum weight retention. With the exception of breastfeeding, all of these factors were also associated with major weight gain. Women under 20 y or over 40 y at delivery, and single women retained significantly more weight. Lower income women with gestational weight gains above the Institute of Medicine (IOM) range retained 3.73 kg more than lower income women who gained within the range. They were also 4.7 times more likely to experience major weight gain with childbearing. The impact of exceeding the IOM gestational weight gain guidelines was three times greater in lower income women than it was in higher income women.

CONCLUSION: Gestational weight gain, postpartum exercise frequency, and food intake are significantly associated with weight change from early pregnancy to 1 y postpartum and major weight gain with childbearing. Lower income women who gain more weight in pregnancy than the IOM recommends are at high risk for major weight gain with childbearing.


Obesity is a major public health concern because of its increasing prevalence and the significant morbidity and mortality associated with excess body weight. Data from the Third National Health and Nutrition Examination Survey (NHANES III) indicate that 23% of women aged 25–55 y are overweight and 26% are obese using the most recent weight classification system for body mass index (BMI).1 Childbearing is associated with weight gains of 0.4–1.8 kg more than those due to aging.2,3,4,5 Although the average amount of weight retained as a result of a pregnancy is modest and many women return to their pre-pregnancy body weights, the amount of weight retained is highly variable and some women retain significant amounts of weight.6,7 For example, Ohlin and Rossner7 report changes in body weight of −12.3 to +26.5 kg from preconception to 1 y postpartum. In various studies, the proportion of women retaining 5 kg or more after 6 months postpartum ranged from 14 to 20%.6,7,8,9 Thus, a subset of women seems to be at increased risk of gaining significant amounts of weight with childbearing.

In a recent review of the literature, Gunderson and Abrams10 state that gestational weight gain is the primary and most important determinant of weight change from preconception to postpartum, but postpartum behaviors are also likely influences on postpartum weight change. Gestational weight gain is strongly related to postpartum weight retention in studies that have used multivariate linear regression analysis.6,7,8,11,12,13 Further, using self-reported weight data from 1988, researchers have shown that women who gain more weight during pregnancy than the Institute of Medicine (IOM) recommends also retain more weight postpartum than women who gain within the IOM range.9,14

In their review, Gunderson and Abrams10 note that few studies have assessed the impact of behavioral factors such as physical activity, dietary intake and lactation on postpartum weight change. Moreover, most of the research found that the associations between these behaviors and weight change were absent or weak. Only four studies have examined some measure of postpartum exercise or physical activity in relation to postpartum weight retention.6,7,11,15 Two6,11 found no relationship and two found that women who engage in physical activity retain less weight at 6 weeks15 and 1 y postpartum.7

Studies examining the relationship between food intake in the postpartum period and weight retention are also limited and show mixed results. A study of WIC participants by Boardley et al11 found no relationship between absolute energy intake in the postpartum period, assessed by a food frequency questionnaire, and weight change from pre- pregnancy. Ohlin and Rossner16 found that women who reported increased food intake during and after pregnancy (first 6 months postpartum) were more likely to retain at least 5 kg at 1 y postpartum compared to women who maintained or decreased their food intake.

The findings on the relationship between breastfeeding and postpartum weight changes are also inconsistent.10,17 Of the six studies examined by Gunderson and Abrams,10 three found no relationship between lactation and postpartum weight retention6,11,14 and the other three found significant relationships but in opposite directions.2,7,8 In one large prospective cohort study, Ohlin and Rossner7 found an effect of lactation on postpartum weight loss in only a small group of women with intensive and lengthy lactation.

Overall, no clear-cut conclusions on the importance of postpartum behaviors for weight retention and major weight gain emerge from this brief review of the literature. Further, none of the studies includes all of these behaviors in combination with gestational weight gain, in examining weight change from early pregnancy to 1 y. Thus the research reported here has two objectives: (1) to describe the relative importance of gestational weight gain, postpartum exercise frequency, food intake, and breastfeeding to weight change from early pregnancy to 1 y postpartum (also referred to as weight retention); and (2) to identify subgroups of women who may be at greatest risk for weight retention and major weight gain with childbearing.

Materials and methods

Study population and sample

This study is a prospective cohort study of 622 healthy adult (≥18 years) women who gave birth to singleton infants and were followed from early pregnancy until 1 y postpartum. Only the 597 women (96%) who delivered full-term (≥37 weeks gestation) infants are included in the sample for the research presented here. The sample was recruited from the population of women who registered for obstetrical care over a 2 y period in a hospital and primary care clinic system serving a 10 county area of upstate New York. The population and sample are primarily white (96%), rural and socio-economically diverse. The participation rate and detailed characteristics of the sample have been described elsewhere.18,19

Data collection

Three methods of data collection were employed. The women were mailed questionnaires and food frequencies at each of the following time periods: first or second trimester of pregnancy, 6 months postpartum, and 1 y postpartum. Shortly after delivery, the women's obstetrical records were audited and the data entered directly into a computer. Eight percent of the medical charts were selected randomly, re-audited by another individual, and they showed excellent agreement with the initial audit. In addition, body weight and height were measured following study protocols (empty bladder, in stocking feet and street clothing) by health care providers at antenatal visits and 1 y postpartum. Weight was measured on Seca digital scales (Columbia, MD, USA). Height was measured using a tape measure fixed to the wall and a moveable head board. Study procedures were approved by the Institutional Review Boards of Bassett Healthcare and Cornell University and were in accord with the Helsinki Declaration of 1975 as revised in 1983.

One year postpartum weights were available for 540 (90%) of the 597 women. Thirty women had no 1 y weight available; 20 had an invalid 1 y weight, primarily because they were pregnant; and seven women were eliminated for one of two reasons—serious postpartum illness affecting body weight or last prenatal weight taken more than 6 weeks before delivery. For 32 of the 540 women (6%), the 1 y postpartum weight used in the analysis was determined from a self-reported body weight recorded on the 1 y questionnaire using the prediction equation from 373 women for whom both a self-reported and measured weight were available (calculated weight (kg)=self-reported weight × 1.003 + 1.006; r2=0.97). Women whose 1 y weights were calculated did not differ significantly on any of the socio-demographic characteristics nor in pre-pregnancy BMI from those with measured weights.

One year questionnaires were returned by 455 of the 540 (84%) women. Seventy of the 455 women completed a shortened version of the questionnaire by telephone interview. These women generally did not complete the food frequency questionnaire.

Determining early pregnancy body weight

Five-hundred and forty-seven of the 597 women (92%) entered the study in the first trimester of pregnancy. Their measured weight at the first prenatal visit in the first trimester of pregnancy was used to calculate gestational weight gain, weight retention, and early pregnancy BMI. For the remaining 8% of the sample who entered the study in the second trimester, their initial weight from this time point was adjusted to 9–11 weeks gestation (the first trimester) using a regression model with variables for the time of measurement in the second trimester (the details of this are available in Olson and Strawderman20).


The outcome variable was weight change, in kilograms, from early pregnancy to 1 y postpartum, also referred to as weight retention. The values for this variable were determined by subtracting the woman's weight at the first prenatal visit in the first trimester of pregnancy from her weight at 1 y postpartum. Major weight gain was defined as retaining 4.55 kg (10 lb) or more at 1 y postpartum.

The IOM body weight categories were used for classifying BMI as underweight (<19.8), normal (19.8–26.0), overweight (26.1–29.0) and obese (>29.0).21 This BMI classification system is the only one currently available that is linked to recommended gestational weight gains.

Gestational weight gain was determined by subtracting the woman's weight at her first prenatal visit in the first trimester from her weight at the last prenatal visit. For the majority of women the last visit occurred within the week before delivery. This difference was then categorized as gaining less than recommended, the recommended amount, or more than recommended according to the IOM gestational weight gain guidelines.21 For obese women, the upper limit of the recommended range was set at 11.5 kg.

For the behavioral risk factors, the overall strategy was to operationalize variables in ways that were both valid and readily implemented in clinical practice. Thus the Godin Leisure-Time Exercise Questionnaire22 was used to assess physical activity. This questionnaire has demonstrated validity and reliability for measuring physical activity in free-living populations.23,24,25 Women were asked about how frequently they participated in regular exercise: ‘How often during your free time do you get regular exercise which makes you sweat or breathe hard?’ The response categories were Never, Rarely, Sometimes and Often=Everyday.

Additional questionnaire items allowed for further validation of this measure in this sample of postpartum women. Two parallel sets of questions asked about the frequency (1–3 times/month, 1–2, 3–4, 5+ times/week) and duration (10–30, 31–60 and>60 min) of leisurely walking and vigorous exercise. Vigorous exercise was defined as activities that caused large increases in breathing, heart rate or leg fatigue or caused sweating and lasted at least 10 min. Energy expended per week was calculated from the reported frequency and duration of walking and vigorous exercise using the midpoint of each response category, body weight at 1 y postpartum, and energy expenditure values of 2.5 and 5.0 METS for walking and vigorous exercise, respectively.25

The measure for food intake was change in the amount of food eaten during the second 6 months compared with the first 6 months postpartum. Women were asked, ‘How has the amount of food you eat now changed compared to the first 6 months since you had your baby?’ The response categories were ‘A lot less food’, ‘A little less food’, ‘About the same’, ‘A little more food’, and ‘A lot more food’. The validity of this question as an accurate measure of change in food intake was assessed using energy intake from the Willet food frequency instrument.26 This food frequency instrument had been used previously by Brown et al27 to detect a change in energy intake from the preconception to prenatal periods and determined to be suited for identifying pregnancy-related changes in energy intake.

Breastfeeding data were collected at the six week obstetrical visit and on the 6 month and 1 y questionnaires. At each time point women were asked if they were currently breastfeeding, if they had ever breastfed and how long they breastfed. Information about introduction of formula was also assessed at 6 weeks and at 6 months postpartum. Breastfeeding after 6 months was assumed to be non- exclusive. A breastfeeding score similar to the one used by Ohlin and Rossner7 was constructed in order to capture duration and intensity of breastfeeding. Each week of exclusive lactation was given 1 point; each week of mixed feeding was given 0.5 point. Missing data for a subject at a particular time point were recovered whenever possible by utilizing breastfeeding information reported at all time points.

Data analysis

Evaluation of independent measures of physical activity and food intake

To test for differences in the total energy expended per week through walking and vigorous exercise across the categories of exercise frequency, the Wilcoxon rank sum test was used. Using the food frequency data, the difference in the daily energy intake from 6 months to 1 y postpartum was calculated for each woman. The Wilcoxon rank sum test was used to assess change in total daily energy intake across the five categories of self-reported behavior change in amount of food intake.

Outcome analysis

Both univariate and multiple linear regression analysis with weight change, in kilograms, at 1 y postpartum as the dependent variable and gestational weight gain category, exercise frequency, change in food intake, and breastfeeding as the independent variables were performed on a personal computer using SAS version 8.0 (SAS Institute, Cary, NC, USA). Although only weights taken between 9 and 19 months postpartum were eligible for inclusion, the actual number of months postpartum at the measurement of the 1 y postpartum body weight was entered into all of the regression models to control for variation in the timing of measurement. Age (in three categories: <20 y; 20–40 y; and>40 y), educational level (<15 y and 15 y or more), smoking status at 1 y postpartum (yes or no), annual household income level (in two categories: ≤185% of the federal poverty index ratio, (PIR) and>185% of the PIR), marital status (single (never married and not living with the baby's father) and married/living with baby's father), pre-pregnancy BMI category (<19.8, 19.8–26.0, 26.1–29,>29), and parity (nulliparous or prima/multiparous) were initially included in the models as potential confounding variables. The categories for each variable were based on capturing variation in the most meaningful way. For example, those women with PIR ≤185% of federal poverty line are income eligible for the expanded Medicaid coverage of prenatal care and are generally considered to have low incomes.

A model reduction (backward elimination) method was applied at the 10% significance level to produce an inclusive, reduced model. In addition, an analysis of model fit was conducted and identified three subjects who were extremely poorly fit in the full model. These subjects were removed from the analysis and the model reduction procedure reapplied. The final model reported here excludes the three outliers. Subjects with missing information regarding factors included in the models were excluded from analysis. All two-way interactions were tested.

In a similar way, major weight retention at 1 y postpartum was analyzed using multiple logistic regression. As a summary of the proportion of variation explained by the model, the max-rescaled r2 was calculated.28 To assess the model's predictive accuracy, the percent of subjects correctly classified was calculated based on cross-validated predicted probabilities. This approach reduces the bias created by using the same data to both fit and assess the model.


Characteristics of women

The socio-demographic and other characteristics of the 540 women in the analysis cohort are shown in Table 1. A majority of the women (50.6%) started pregnancy with a normal body mass index (BMI). The second largest BMI group was obese (25.2%). The majority of women in the sample was between 20 and 40 y of age (93.5%) and was married or living with a partner (92.8%). The sample was distributed fairly equally across the two categories for each of the following variables: educational level, annual household income, and parity.

Table 1: Characteristics of the sample (n=540)

Gestational weight gain and behavioral risk factors

Only about 38% of women gained an amount of weight in pregnancy that is within the guidelines recommended by the IOM (Table 2). An even larger proportion (42%) gained more than the IOM recommended amount. While frequent exercise was not common, 11% of women exercised daily and an additional 32% exercised ‘Sometimes’. During the second 6 months postpartum most women decreased (49.4%) their food intake, with 13% reporting they decreased their food intake ‘A lot’ (Table 2). The number of women who reported eating ‘A lot more food’ was small (n=8) so these women were combined with the women who reported eating ‘A little more food’ to create a ‘More food’ group consisting of 9.1% of the sample for use in the weight retention analysis.

Table 2: Risk factors for weight change from early pregnancy to 1 y postpartum: distribution of the sample across categories and relationship to weight retention (kg) at 1 y postpartuma

Twenty-five percent of women were breastfeeding at the 1 y postpartum. The prevalence of breastfeeding in the sample was generally high: at the routine 6 week postpartum visit, 81% of women reported they had initiated breastfeeding at birth and 66% reported they were still breastfeeding. Forty-eight percent of women reported that they were breastfeeding at 6 months postpartum.

Validity of physical activity and food intake measures

The overall mean (±s.e.m.) energy expended per week during leisurely walking and vigorous exercise was 1691.1±104.7 kJ. Mean weekly energy expenditure significantly declined with decreasing frequency of exercise. Women who exercised ‘Often’ expended significantly more energy (P<0.001) than those who exercised ‘Sometimes’ (4165.5±411.5 vs 1811.3±138.6 kJ). Those exercising ‘Sometimes’ expended significantly more energy (P< 0.0001) than those who rarely exercised (801.2±74.9 kJ). Finally, those who rarely exercised expended more (P= 0.005) energy relative to those who never exercised (540.4±139.8 kJ).

The mean (±s.e.m.) change in energy intake during the second 6 months postpartum for the sample was −318.1± 129.8 kJ/day. The mean change in daily energy intake for women who reported their food intake was a lot less relative to the first 6 months postpartum was −1197.2±373.4 kJ. This change was significantly greater (P=0.01) than the change in the other groups. Women reporting a small decrease, no change or an increase in the amount of food consumed had a mean change in energy intake of −209.3±138.1 kJ. No significant differences in energy intake were detected between these latter groups. Thus questions about the exercise frequency and change in the amount of food intake appear to be measuring changes in energy expenditure and energy intake, respectively.

Weight change from early pregnancy to 1 y postpartum

At 1 y postpartum, women were on average 1.51±5.95 kg heavier than they were in early pregnancy. Postpartum weight retention was highly variable, ranging from −19.09 to 27.50 kg. Some 41.7% of the women had body weights that were equal to or less than their early pregnancy body weights, 32.8% retained 0–4.55 kg (10 lb) and 25.6% were at least 4.55 kg heavier at 1 y postpartum than they were in early pregnancy.

The results of the univariate or simple regression analyses examining the relationships of each of the behavioral risk factors (exercise frequency, change in food intake, and breastfeeding at 1 y) to weight retention at 1 y are shown in Table 2. Women who exercised often (P<0.001), ate a lot less food (P=0.04), and were breastfeeding at 1 y (P=0.04) retained significantly less weight compared with their corresponding reference groups. Women who ate more retained significantly more weight (P=0.006) than those who ate the same amount of food. Breastfeeding at all other times points and the breastfeeding score were not significantly related to postpartum weight retention.

Figure 1 shows the adjusted mean weight retention for each of the gestational weight gain categories of women in each of the early pregnancy BMI groups. All values are relative to normal BMI women gaining within the IOM guidelines. Women in the normal, high and obese BMI groups who gained more weight in pregnancy than is recommended by the IOM retained significantly more weight at 1 y (all P<0.01). High BMI women who gained less than the recommended amount were 4.31 kg lighter than normal BMI women who gained the recommended amount (P=0.05).

Figure 1
Figure 1

Weight change at 1 y postpartum, in kg, by initial body mass index (BMI) category and gestational weight gain (GWG) group (adjusted for number of months postpartum and relative to normal BMI and recommended GWG). Model F=4.52; P≤0.0001; adjusted r2=0.07; n=540.

Table 3 shows the final multivariate regression model controlling for potential selection bias and confounding variables. Women who ate much less food (P=0.008), gained less weight in pregnancy than the IOM recommends (P=0.016), and breastfed until 1 y postpartum (P=0.02) retained significantly less weight than women who maintained or increased their food intake, gained the recommended amount or more during pregnancy, and breastfed until one year postpartum.

Table 3: Multivariate regression model of risk factors (and significant interactions) for weight change from early pregnancy to 1 y postpartum (weight retention, in kg)a

Several of the potential confounding variables were also significant. Women who were under 20 y of age, over 40 y, or single, not living with the baby's father, retained significantly (P<0.05) more weight at 1 y postpartum.

Two significant interactions were found between the risk factors and the characteristics of women: exercise frequency with BMI and gestational weight gain with income. These are illustrated in Figure 2, panels A and B showing the unadjusted means and standard errors for each group. Within the group of normal and low BMI women, exercising often was associated with significantly less weight retention (−1.80 kg; P=0.03), but within the group of women with BMI greater than normal, exercising often was associated with even less weight retention (−1.80+(−3.61)=−5.41 kg; P=0.006; see Table 3.) Gaining more than the IOM recommends during pregnancy was not significantly related to weight retention at 1 y in women with annual household incomes greater than 185% PIR. However, lower income women who gained more in pregnancy than the IOM recommends were 3.73 kg heavier at 1 y postpartum than lower income women who gained below or within the IOM guidelines for gestational weight gain (P<0.001).

Figure 2
Figure 2

Unadjusted mean weight change from early pregnancy to 1 y postpartum (weight retention) for variables with significant interactions in Table 3.

Holding gestational weight gain constant at the level of gaining within the IOM guidelines, a significant interaction was also found between two characteristics of the women, BMI and annual household income. (Illustrated in Figure 2, panel C.) Lower income women (annual family income of<185% PIR) with greater than normal BMI retained 2.28 kg more than lower income women with normal or low BMI. However, lower income women with low or normal BMI and gestational weight gain equal to or less than the IOM recommends tended to retain less weight at 1 y postpartum, although this was not statistically significant.

Risk of major weight gain

During the period from pregnancy to 1 y postpartum about a quarter of the sample experienced major weight gain (retained 4.55 kg/10 lb or more). Exercising often and gaining less weight in pregnancy than is recommended by the IOM were significantly (P<0.05) associated with decreased risk of major weight gain (Table 4). Increasing food intake in the second 6 month postpartum and being less than 20 y old at delivery were significantly associated (P<0.05) with greater risk of major weight gain. Gaining more weight in pregnancy than is recommended by the IOM, by itself, was not significantly associated with major weight gain (P=0.20), but interaction of this variable with income was significant (P<0.05). As shown in Figure 3, the lower income women who exceeded the IOM guidelines were twice as likely to retain 4.55 kg or more than higher income women who exceeded the IOM guidelines. The risk of retaining 4.55 kg or more for those who gained the recommended amount or less did not differ by income group.

Table 4: Odds ratios for major weight gain (≥4.55 kg or 10 lb) at 1 y postpartuma
Figure 3
Figure 3

Proportion of women with major weight gain (≥4.55 kg or 10 lb) at 1 y postpartum by gestational weight gain and income group.


The amount of weight retained at 1 y postpartum in this sample (mean=1.51 kg; s.d.=5.95 kg) is similar to that found in the other two cohort studies. Schauberger et al6 found an average retention of 1.4 kg at 6 months postpartum in their study of 795 white middle-class women. Ohlin and Rossner7 found an average retention of 1.5 kg at 1 y postpartum in 1423 Swedish women. Gunderson and Abrams,10 who considered the 10 studies in the literature with study designs that would support valid estimation, found very similar means and standard deviations to those above.

Gestational weight gain, frequency of exercise, change in the amount of food eaten, and breastfeeding at 1 y postpartum are significantly related to weight retention in the expected directions. The effects of the behavioral risk factors are independent of each other and are therefore additive. Using Table 3 as a source, and considering a normal or low BMI woman with an annual household income of greater than 185% PIR, the sum of the parameter estimates for these four risk factors is 5.13 kg or approximately 11.3 lb, an amount that would be considered a major weight gain by most women and healthcare providers. The proportion of women with any one of these risk factors is large, ranging from 42% for excessive gestational weight gain to 85% or greater for not sufficiently reducing food intake or lack of daily exercise. Furthermore, a substantial proportion of women have more than one of these risk factors. Twenty-three percent of women have all four, 46% of women have three, and 25% have two risk factors. While previous studies reviewed by Gunderson and Abrams10 have examined one or two of these risk factors at a time, this is one of the first studies to examine them simultaneously. The results of this more complete picture indicate that these risk factors have a substantial impact on weight change from early pregnancy to 1 y postpartum and risk of major weight gain.

The results also indicate that the importance of these risk factors varies substantially for subgroups of women. For example, gestational weight gain is a very important risk factor for both postpartum weight retention and the risk of major weight gain in lower income women. For women with household incomes greater than 185% PIR and normal or low BMI, gestational weight gain is only of minor importance and the behavioral risk factors of exercise frequency, change in food intake, and breastfeeding appear more important.

Confidence in the findings presented here is strengthened by several of the study's design features and methods. The study has a prospective cohort design and the combined drop-out and lost-to-follow up rate was 5%. Ninety-five percent of the women who entered the study remained in the study at 1 y postpartum, in comparison to the approximate 60% of women at the end of the follow-up period in other cohort studies.6,7 The study population was confined to women who were at least 18 y of age at the time of delivery and thus maternal growth is not a likely contributor to weight gain in this study. Finally, with an analytical sample size of 450–540 women, the study had adequate statistical power to demonstrate a significant relationship between the independent variables and postpartum weight retention.

In addition, measured rather than self-reported weights were used at all time points for the vast majority of women, 88% for the initial weight and 94% for the 1 y postpartum weight. Accepted methods were used for estimating measured weights from self-reported weights for the 30 women with missing 1 y measured weights.

Valid, yet relatively simple, behavioral measures of exercise and food intake were used. Results presented in the paper support the use of these measures as accurate indicators of energy expenditure and energy intake. The reported changes in behavior appear to be intentional as they are correlated with women's answers to questions about doing anything to lose weight. Chi-square analysis indicated significant associations (P<0.001) for both exercise frequency and food intake. For example, 65% of the women who reported they exercised often reported they were exercising to lose weight vs 31% of women who exercised less often. Similar results were found for the relationship between eating less and report of eating less food to lose weight.

Some might argue that the outcome variables in this study (weight in kg retained at 1 y and major postpartum weight retention of 4.55 kg or more) suffer from the statistical issue of part–whole bias as discussed by Gunderson and Abrams10 and Selvin and Abrams29 in relation to maternal gestational weight gain and infant birth weight. Part–whole bias is the inflated correlation between an outcome and a predictor when the predictor is a component of the outcome. In this study, both weight retention at 1 y and gestational weight gain use the early pregnancy weight as a baseline, making gestational weight gain a mathematical component of the weight retention. Some investigators choose to model gestational weight gain as a predictor of weight change from delivery until some postpartum time point. However, this outcome, although avoiding the part–whole bias, is not informed by pre- or early pregnancy body weight. This weight adds substantial meaning to the values for both delivery and any postpartum weights. To illustrate, two different women could have delivery weights of 70 kg, but one woman weighed 70 kg in early pregnancy and another weighed 55 kg. The meaning of a 2 kg weight loss from delivery to 1 y postpartum would be very different for these two women. Thus to minimize any inflated association between gestational weight gain and weight retention and to also maintain consideration of early pregnancy weight, we used early pregnancy weight in the calculation of the outcome and used a categorical measure of gestational weight gain.

A limitation of the study is that the population of women was primarily white despite having diverse socio-economic and educational backgrounds. The relationship between gestational weight gain, behavioral risk factors, and postpartum weight retention needs to be investigated in a more racially and ethnically diverse population. Further, while this study included a fairly large number of variables, we may have failed to include important predictor variables. The model shown in Table 4 correctly classifies 79% of the 446 subjects with respect to major weight gain category. However, slightly less than 50% of the women who experience major weight gain are correctly classified, indicating the model is less sensitive than would be desired for clinical application. Thus there is a need for more research aimed at refining the measurement of the current risk factors and identifying additional risk factors for postpartum weight gain.

The work reported here sits at the intersection of two important public health issues: healthy birth weights in infants and chronic disease prevention for childbearing women. A growing body of literature supports the appropriateness of the IOM gestational weight gain guidelines for achieving optimal birth weights.30,31,32 The findings presented here support the appropriateness of the IOM guidelines for preventing major weight retention associated with childbearing.

The women in the ‘Often’ exercise group were expending an amount of energy that was roughly equivalent to the current public health recommendation of 30 min or more of moderate-intensity physical activity on most, preferably all, days of the week.33 Several studies in the early postpartum period have demonstrated that this level of energy expenditure has no untoward consequences for the infants of breastfeeding mothers.34,35,36

While findings indicate that continuing breastfeeding to 1 y postpartum may benefit women in terms of postpartum weight retention, breastfeeding does not appear to be a major avenue for minimizing it. Since any effect will not be apparent until well after the initiation of lactation and given other more immediate benefits of breastfeeding to maternal and infant health, promotion of breastfeeding should highlight these benefits rather than minimizing weight retention.

In conclusion, the results reported here indicate that the ranges for gestational weight gain outlined in the IOM reports Nutrition during pregnancy21 and Nutrition during pregnancy and lactation: an implementation guide37 are appropriate for promoting healthy body weights and preventing major weight gain in childbearing women. Given the well- documented association between the advised and actual gestational weight gains,38,39 health care providers can play an important role in preventing the development of obesity associated with childbearing by using the IOM gestational weight gain guidelines in the delivery of prenatal care. This is especially important for lower income women such as those enrolled in the Special Supplemental Nutrition Program for Women, Infants and Children (WIC). The other behavioral risk factors, postpartum exercise frequency and decreased food intake, are independent of and their effects are additive to those of gestational weight gain. They open a second opportunity for preventing major weight gain with childbearing.


  1. 1.

    , & . The disease burden associated with overweight and obesity. JAMA 1999; 282: 1523–1529.

  2. 2.

    , , & . The effect of pregnancy on the body mass index 9 months postpartum in 49 women. Int J Obes 1987; 11: 609–618.

  3. 3.

    , , , & . Parity-associated weight gain and its modification by sociodemographic and behavioral factors: a prospective analysis in US women. Int J Obes Relat Metab Disord 1997; 21: 802–810.

  4. 4.

    , , , , & . A prospective study of childbearing and 10-year weight gain in US white women 25 to 45 years of age. Int J Obes Relat Metab Disord 1994; 18: 834–838.

  5. 5.

    , , , , & . Longitudinal changes in adiposity associated with pregnancy. The CARDIA Study. Coronary Artery Risk Development in Young Adults Study. JAMA 1994; 271: 1747–1751.

  6. 6.

    , & . Factors that influence weight loss in the puerperium. Obstet Gynecol 1992; 79: 424–429.

  7. 7.

    & . Maternal body weight development after pregnancy. Int J Obes 1990; 14: 159–173.

  8. 8.

    , , & . Postpartum weight change: how much of the weight gained in pregnancy will be lost after delivery. Obstet Gynecol 1988; 71: 701–707.

  9. 9.

    & . Pregnancy-related weight gain and retention: Implications of the 1990 Institute of Medicine guidelines. Am J Public Health 1993; 83: 1100–1103.

  10. 10.

    & . Epidemiology of gestational weight gain and body weight changes after pregnancy. Epidemiol Rev 1999; 21: 261–275.

  11. 11.

    , , , & . The relationship between diet, activity, and other factors, and postpartum weight change by race. Obstet Gynecol 1995; 86: 834–838.

  12. 12.

    , & . Timing of weight gain during pregnancy: promoting fetal growth and minimizing maternal weight retention. Int J Obes Relat Metab Disord 1996; 20: 526–532.

  13. 13.

    , , , & . Gestational weight gain, pregnancy outcome, and postpartum weight retention. Obstet Gynecol 1995; 86: 423–427.

  14. 14.

    & . Differences in postpartum weight retention between black and white mothers. Obstet Gynecol 1993; 81: 768–774.

  15. 15.

    , , , & . Physical activity and postpartum well-being. J Gynecol Obstet Nurs 1999; 28: 41–49.

  16. 16.

    & . Trends in eating patterns, physical activity and sociodemographic factors in relation to postpartum body weight development. Br J Nutr 1994; 71: 457–470.

  17. 17.

    . Weight changes during pregnancy and the postpartum period. Prog Food Nutr Sci 1991; 15: 117–157.

  18. 18.

    & . Predictors of pregnancy-associated change in physical activity in a rural white population. Matern Child Health J 2001; 5: 7–14.

  19. 19.

    , & . Evaluation of psychosocial measures for understanding weight-related behaviors in pregnant women. Ann Behav Med 2001; 23: 50–58.

  20. 20.

    & . Modifiable behavioral factors in a biopsychosocial model predict inadequate and excessive gestational weight gain. J Am Diet Assoc, (in press).

  21. 21.

    Institute of Medicine . Nutrition during pregnancy. Part I, Weight gain, National Academy Press: Washington, DC; 1990; p 10

  22. 22.

    & . A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 1985; 10: 141–146.

  23. 23.

    , , & . A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med Sci Sports Exerc 1993; 25: 81–91.

  24. 24.

    , , , , & et al . A collection of physical activity questionnaires for health-related research. Med Sci Sports Exerc 1997; 29(Suppl 6): S36–S38.

  25. 25.

    , , , , & et al . Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993; 25: 71–80.

  26. 26.

    , & . Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 1985; 122: 51–65.

  27. 27.

    , , , , & et al . A food frequency questionnaire can detect pregnancy-related changes in diet. J Am Diet Assoc 1996; 96: 262–266.

  28. 28.

    . A note on a general definition of the coefficient of determination. Biometrika 1991; 78(3): 691–692.

  29. 29.

    & . Analysing the relationship between maternal weight gain and birthweight: Exploration of four statistical issues. Paediatr Perinat Epidemiol 1996; 10: 220–224.

  30. 30.

    & . Prenatal weight gain advice: an examination of the recent prenatal weight gain recommendations of the Institute of Medicine. Obstet Gynecol 1992; 79: 664–669.

  31. 31.

    , , & . Gestational weight gain among average-weight and overweight women—what is excessive? Am J Obstet Gynecol 1995; 172: 705–712.

  32. 32.

    , & . An empiric evaluation of the Institute of Medicine's pregnancy weight gain guidelines by race. Obstet Gynecol 1998; 91: 878–884.

  33. 33.

    , , , , & et al . Physical activity and public health—a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA 1995; 273: 402–407.

  34. 34.

    , , , & . Randomized trial of the short-term effects of dieting compared with dieting and aerobic exercise on lactation performance. Am J Clin Nutr 1999; 69: 959–967.

  35. 35.

    , , & . The effect of weight loss in overweight, lactating women on the growth of their infants. New Engl J Med 2000; 342: 449–453.

  36. 36.

    , & . Is milk production impaired by dieting during lactation. Am J Clin Nutr 1994; 59: 833–840.

  37. 37.

    Institute of Medicine . Nutrition during pregnancy and lactation: an implementation guide, National Academy Press: Washington, DC; 1992.

  38. 38.

    , & . Medical advice on maternal weight gain and actual weight gain. Ann NY Acad Sci 1993; 678: 293–305.

  39. 39.

    , , & . Medically advised, mother's personal target, and actual weight gain during pregnancy. Obstet Gynecol 1999; 94: 616–622.

Download references


This research was supported by NIH grants HD29549 and HD07331.

Author information


  1. Division of Nutritional Sciences, Cornell University, Ithaca, New York, USA

    • C M Olson
  2. Department of Food Science and Human Nutrition, University of Missouri, Missouri, USA

    • M S Strawderman
  3. Department of Community and Preventive Medicine, University of Rochester School of Medicine and Dentistry, Rochester, New York, USA

    • P S Hinton
    •  & T A Pearson


  1. Search for C M Olson in:

  2. Search for M S Strawderman in:

  3. Search for P S Hinton in:

  4. Search for T A Pearson in:

Corresponding author

Correspondence to C M Olson.

About this article

Publication history






Further reading