International Journal of Obesity (2002) 26, 1494–1502
ß 2002 Nature Publishing Group All rights reserved 0307–0565/02 $25.00
www.nature.com/ijo
PAPER
Randomized controlled trial to prevent excessive
weight gain in pregnant women
BA Polley1, RR Wing1* and CJ Sims2
1
University of Pittsburgh, Pittsburgh, Pennsylvania, USA; and 2Magee Womens Hospital, Pittsburgh, Pennsylvania, USA
BACKGROUND: The Institute of Medicine (IOM) recommends that normal-weight women (BMI (body mass index) of
19.8 – 26.0) gain 25 – 35 lb (11.4 – 15.9 kg) during pregnancy, and that overweight women (BMI of 26.1 – 29.0) gain 15 –
25 lbs (6.8 – 11.4 kg). A significant number of normal-weight women and an even greater proportion of overweight women
exceed these guidelines, which increases postpartum weight retention and may contribute to the development of obesity.
OBJECTIVE: To determine whether a stepped care, behavioral intervention will decrease the percentage of women who gain
more than the IOM recommendation.
DESIGN: Randomized controlled trial comparing a stepped-care behavioral intervention with usual care. Women (n ¼ 120) who
had a BMI > 19.8, age > 18 and < 20 weeks gestation were recruited from a hospital-based clinic serving low-income women
and randomized by race and BMI category to the intervention or control group. The intervention group received education
about weight gain, healthy eating, and exercise and individual graphs of their weight gain. Those exceeding weight gain goals
were given more intensive intervention. Women were followed through pregnancy to their first postpartum clinic visit. The main
outcome measure was weight gain during pregnancy categorized as above the IOM recommendations vs below or within the
IOM recommendations.
RESULTS: The intervention significantly decreased the percentage of normal-weight women who exceeded the IOM
recommendations (33 vs 58%, P < 0.05). There was a non-significant (P ¼ 0.09) effect in the opposite direction among
overweight women (59% of intervention and 32% of control gained more than recommended). Postpartum weight retention
was strongly related to weight gain during pregnancy (r ¼ 0.89).
CONCLUSIONS: The intervention reduced excessive weight gain during pregnancy among normal weight women.
International Journal of Obesity (2002) 26, 1494 – 1502. doi:10.1038=si.ijo.0802130
Keywords: pregnancy; prevention of obesity; weight gain; randomized controlled trial; behavioral intervention
Introduction
The Institute of Medicine (IOM) recommends that normalweight women (BMI (body mass index) 19.8 – 26.0 kg=m2)
gain 25 – 35 lb (11.4 – 15.9 kg) during pregnancy, and that
overweight women (BMI 26.1 – 29.0 kg=m2) gain 15 – 25 lb
(6.8 – 11.4 kg). Weight gains within these guidelines are
associated with healthy fetal and maternal outcomes;
weight gains below these goals are associated with low
infant birthweight and higher weight gains are associated
with macrosomia in the infant.1,2 Moreover, women who
gain more than recommended retain twice as much
weight after pregnancy as women who gain within the
*Correspondence: RR Wing, Brown University=Miriam Hospital, Weight
Control and Diabetes Research Center, 14 Third Street, Providence,
RI 02906, USA.
E-mail: rwing@lifespan.org
Received 10 January 2002; revised 30 April 2002;
accepted 24 May 2002
recommendations.3,4 For example, in the National and
Maternal Health Survey, white women who gained above
the IOM recommended level retained 2.2 kg at 10 – 18
months postpartum, whereas those who gained within the
IOM recommendation retained only 0.7 kg. African American women who exceeded the recommended level of
weight gain retained 5.8 kg, whereas those who gained
within the recommended levels retained 3.3 kg.3 Weight
gain during pregnancy may thus contribute to the development of obesity in young women.
Although current weight gain goals are more liberal than
in the past, a significant proportion of pregnant women
exceed these current goals.3,5 Based on data from the
National Maternal and Infant Health Survey, which includes
a representative sample of women in the United States who
had live births in 1988, 36% of women exceed the IOM
recommendations. Only 20% of underweight women
exceeded these goals, but 37% of normal-weight women
Prevention of excessive weight gain in pregnancy
BA Polley et al
1495
and 64% of overweight women gained more than recommended.3 Similarly, in a study of over 4000 women from the
University of California, San Francisco Perinatal Database,
23% of underweight women, 49% of normal-weight women,
and 70% of overweight women exceeded the recommendations.5 Thus a substantial proportion of normal-weight
women and an even greater proportion of overweight
women gain more than is recommended.
Weight gain during the first 20 weeks of gestation has
been shown to be predictive of total weight gain;6 – 8 thus, it
may be possible to identify women early in pregnancy who
will exceed the weight gain guidelines and counsel them
about healthy eating and exercise habits to prevent excessive
weight gain during pregnancy. This approach fits with the
Subcommittee on Weight Gain of the National Academy of
Science’s recommendation that deviations from the desired
rate of weight gain be discussed with the woman and
corrective actions such as ‘moderating consumption and
increasing physical activity’ be initiated, as appropriate.1
However, no studies to date have examined the efficacy of
providing education and behavioral strategies related to
healthy eating, moderate exercise and appropriate weight
gain goals as a means of helping women achieve appropriate
rates of pregnancy weight gain.
The goal of the present study was to develop and test a
stepped care behavioral intervention to reduce excessive
weight gain during pregnancy. The intervention provided
education and feedback about weight gain, and stressed
modest exercise and healthy, low-fat eating. We examined
whether, compared with a usual care control group, the
intervention would reduce the number of women who
exceed the IOM recommendations. Because normal-weight
and overweight women have different IOM weight gain
goals and appear to differ in the proportion of women
exceeding these goals, we examined the efficacy of the
intervention for these two groups.
All women signed a consent form approved by the Institutional Review Boards of the University of Pittsburgh Medical
Center and Magee-Womens Hospital. Of the 120 women who
entered the study, 110 (92%) were followed through the end
of their pregnancy; three of the 120 women miscarried, six
moved out of the area, and one asked to withdraw from the
study. An additional five in the intervention group requested
withdrawal from the study but gave consent for continued
monitoring via record review and were retained for all analyses. The final sample of 110 women included 57 in the
treatment group (30 normal weight, 27 overweight) and 53 in
the control group (31 normal weight, 22 overweight). Analysis of variance and chi-square showed no differences between
the treatment and control group on any of the demographic
characteristics summarized in Table 1.
Design and procedures
Women were randomly assigned to the standard care control
group or to the intervention stratified by BMI category
(normal weight vs overweight) and self-reported race (black
vs white). BMI was based on retrospective self-reported
height and weight at last menstrual period with normal
weight defined as BMI ¼ 19.8 to 26.0 and overweight
BMI > 26.0. Although the IOM recommends a weight gain
of 6.8 – 11.3 kg for overweight women (BMI of 26.1 – 29) and
a weight gain of 6.8 kg (with no specified upper limit) for
obese women (BMI > 29), these two groups were collapsed in
the present study and the recommendations for overweight
women were used, as occurs in other research on this topic.5
Standard care
Subjects in the standard prenatal care condition received
only the standard nutrition counseling provided by the
Table 1 Demographic characteristics
Methods
Participants
Pregnant women were recruited before 20 weeks gestation
(mean s.d. ¼ 14.5 3.1 weeks) from an obstetric clinic for
low-income women at a hospital in Pittsburgh, PA, USA.
Subjects were recruited into four cells: normal and overweight, black and white. Underweight women (BMI < 19.8
based on self-reported weight at height at the last menstrual
period) were excluded, as were women younger than 18 y,
those whose first prenatal visit was > 12 weeks gestation, and
those with a high-risk pregnancy (ie drug abuse, chronic
health problems, previous complications during pregnancy,
or current multiple gestation). Of 164 women who were
approached, 120 enrolled in the study. Refusal rates were
higher among black women (28=74 refused) than among
white women (16=90 refused), P < 0.005, and higher in overweight blacks (16=37 refused) than in any of the other three
weight-by-race categories (P < 0.02).
Variable
Distribution
Age (y)
Mean ¼ 25.5 4.8
Race
39% black
61% white
Marital status
31% married
14% separated, divorced or widowed
55% never married
Education
45% high school or less
42% vocational training or some college
9% college graduate
4% graduate training
Employment
57% unemployed
14% trade, service
16% clerical, sales
6% professional
7% other
Childbearing history
47% first pregnancy
30% second pregnancy
17% third pregnancy
6% > third pregnancy
International Journal of Obesity
Prevention of excessive weight gain in pregnancy
BA Polley et al
1496
physicians, nurses, nutritionists and WIC counselors at
Magee-Womens Hospital. This counseling emphasized a
well-balanced dietary intake and advice to take a multivitamin=iron supplement. No information or counseling was
provided by the research staff.
Intervention
The intervention was delivered at regularly scheduled clinic
visits by master’s and doctoral level staff with training in
nutrition or clinical psychology. Shortly after recruitment,
women in the intervention condition were given written and
oral information in the following areas: (a) appropriate
weight gain during pregnancy; (b) exercise during pregnancy;
and (c) healthful eating during pregnancy. Newsletters
prompting healthy eating and exercise habits were mailed
biweekly. In addition, after each clinic visit, women in the
intervention condition were sent a personalized graph of
their weight gain. Women whose weight changes were
within the appropriate ranges were informed that they were
gaining the expected amount of weight and were encouraged
to eat a healthy diet and to have a physically active lifestyle;
those gaining too little were told to check with their physician for guidance outside the scope of this study. Women
whose weight gains exceeded the recommended levels were
given additional individualized nutrition and behavioral
counseling using the format listed in Table 2. A steppedcare approach was used, where the woman was given increasingly structured behavioral goals at each visit if her weight
continued to exceed the recommended levels. The primary
focus of the dietary intervention was on decreasing high-fat
foods, such as fast food items, and substituting healthier
alternatives (fruit and vegetables); if these approaches did
not help the woman achieve the IOM recommended weights,
then a more structured meal plan and individualized calorie
goals were added. The exercise intervention focused on
increasing walking and developing a more active lifestyle
(eg walking rather than driving short distances). Between
clinic visits women were contacted by telephone to discuss
progress towards the goals set at the previous visit.
Measures
Pre-pregnancy weight was the weight at the last menstrual
period, as reported by the participant at the time of recruit-
Table 2 Format of individual counseling sessions
Review of weight gain chart
Assessment of current eating and exercise (ie a 24 h recall or examination
of self-monitoring records), with periodic computerized nutrition analysis
Review of progress toward behavioral goals
Problem-solving
Instruction in the use of behavioral techniques such as stimulus control or
self-monitoring
Goal-setting for eating and exercise behaviors
International Journal of Obesity
ment into this study. Weight was measured at every clinic
visit using a digital or balance beam scale. Total weight gain
was based on self-reported pre-pregnancy weight and weight
at the last clinic visit prior to delivery. For five women, the
last clinic visit was more than one month prior to delivery
and self-reported weight at delivery was used. Postpartum
weight was obtained at the first clinic visit after delivery
(mean s.d. ¼ 8 7.1 weeks); self-reported weight was used
for 14 women; postpartum weight was unavailable for 36
women.
Demographic and weight history information was
obtained at recruitment. Self-report measures of dietary
intake and exercise expenditure were obtained at recruitment, 30 weeks gestation, and 6 weeks postpartum. A short
form of the Block Food Frequency Questionnaire,9 administered by interview, was used to assess consumption of 13
major contributors to fat intake over the past month. This
questionnaire was included to examine changes in intake of
high-fat foods rather than to try to describe the participants’
overall dietary consumption. Energy expenditure was measured using the Paffenbarger Exercise Questionnaire10 by
interview, assessing stair-climbing, walking and recreational
activities over the last week.
Obstetric records were reviewed after delivery to obtain
infant birth weight and complications during pregnancy
and=or delivery.
Statistical analysis
The primary dependent variable was the proportion of
women who exceeded the IOM recommendation. This
was analyzed by logistic regression examining the effect
of treatment group and BMI category on weight gain
category (above IOM recommended level vs below IOM
recommended level). Weeks of gestation at delivery was
used as a covariate. Secondary aims were to assess the
effects of treatment group and BMI category on total
weight gain, weight gain from recruitment to delivery,
postpartum weight loss and weight retention. Analysis of
variance was used for these secondary aims; again weeks of
gestation at delivery was used as a covariate. When significant interactions with weight category were observed,
post hoc analyses were conducted to analyze the effects of
treatment separately in normal-weight and overweight
women. Because pre-pregnant weight differed in normalweight women in the control and intervention groups, prepregnancy weight was covaried in these post hoc analyses.
Analyses were done using an intent-to-treat approach,
including all participants available at the post-pregnancy
or postpartum assessments (Figure 1). Both SAS and SPSS
statistical packages were used.
Results
The flow of participants through the study is shown in
Figure 1. Weight change variables are summarized in Table 3.
Prevention of excessive weight gain in pregnancy
BA Polley et al
1497
Figure 1 Flow of subjects.
Excessive weight gain during pregnancy
Initial analyses examined whether women met weight gain
goals at any time during the pregnancy (ie whether the
women’s weight gain ever exceeded the expected range),
and whether they later ‘recovered’ (ie achieved weight
gains within the expected range). As shown in Table 3,
there was a significant interaction (P ¼ 0.01) between treatment group and BMI category for the percentage of women
Table 3
who exceeded the IOM recommendations at some point
during their pregnancy. Among normal-weight women,
94% of the control group exceeded the recommended level
at some point during pregnancy compared with 63% of
women in the intervention group, P < 0.05. Among overweight women, about two-thirds exceeded the recommended level at some point in their pregnancy regardless
of treatment group, P > 0.4.
Weight changes during pregnancy by treatment group and BMI category, mean s.d. or n (%)
Normal-weight
Overweight
Control
(n ¼ 31)
Intervention
(n ¼ 30)
Control
(n ¼ 22)
Intervention
(n ¼ 27)
22.5 2.0
59 6.9
22.8 1.9
62.1 6.1*
34.1 7.2
91.8 19.6
31.4 6.0
83.6 16.7b,c
Exceeded IOM recommendations at
some point during pregnancy, n (%)
Recovery rates, n (%)
29 (93.5%)
19 (63.3%)*
14 (63.6%)
20 (74.1%)
11 (37.9%)
9 (47.4%)
7 (50.0%)
4 (20.0%)
Total weight gain (kg)
(range, kg)
16.4 4.8
(6.8 – 30.9)
15.4 7.1
(2.7 – 32.7)
10.1 6.2
(70.9 – 26.4)
13.6 7.2b
(1.4 – 29.1)
IOM categories of total weight gain
Exceed IOM recommend, n (%)
Mean gain (kg)
Within IOM recommend, n (%)
Mean gain (kg)
Below IOM recommend, n (%)
Mean gain (kg)
18 (58.1%)
(19.1 kg)
8 (25.8%)
(13.6 kg)
5 (16.1%)
(9.5 kg)
10 (33.3%)*
(23.6 kg)
11 (36.7%)
(13.6 kg)
9 (30.0%)
(8.2 kg)
7 (31.8%)
(17.3 kg)
8 (36.4%)
(10 kg)
7 (31.8%)
(3.6 kg)
16 (59.3%)c
(17.7 kg)
6 (22.2%)
(10.5 kg)
5 (18.5%)
(4.1 kg)
Postpartum weight lossa
Net weight retention, kga
n ¼ 21
79.0 2.4
6.2 4.5
n ¼ 18
79.6 3.5*
4.4 5.4
2
Prepregnancy BMI (kg=m )
Prepregnancy weight (kg)
a
n ¼ 19
79.6 2.7
0.3 7.0
b
c
n ¼ 16
79.1 4.1
3.6 5.6
b
Postpartum weights available for 74 participants and all postpartum weights adjusted for weeks postpartum; main effect of
BMI category, P < 0.001; cinteraction between treatment group and BMI category, P < 0.01.
*Indicates a significant treatment effect for post hoc analyses within each BMI category.
International Journal of Obesity
Prevention of excessive weight gain in pregnancy
BA Polley et al
1498
Women in the intervention group who exceeded the
weight gain pattern were notified that they were gaining
above the IOM recommended levels and were provided with
at least one additional counseling session. The number of
additional sessions ranged from 1 to 11, depending on
whether the women’s weight gain ‘recovered’ or remained
above the guidelines. Of the intervention women who
received additional sessions, normal-weight women
received an average of 3.3 additional sessions and overweight women received an average of 4.8 additional sessions. As shown in Table 3, there was a trend for a
significant treatment group-by-BMI category interaction in
the proportion of women who were able to recover from
excessive weight gain during pregnancy and go on to have a
total weight gain within or below IOM recommendations,
P ¼ 0.07. However, there was no significant effect of treatment when normal-weight and overweight women were
analyzed separately (P > 0.1).
Excessive total weight gain
The primary goal of the intervention was to decrease the
percentage of women who exceeded the IOM guidelines
from pre-pregnancy to the end of pregnancy. As shown in
Figure 2, there was a significant interaction between treatment group and BMI category for excessive weight gain,
P < 0.01. Among normal-weight women, those in the intervention group were significantly less likely than controls to
gain above the IOM recommendations, P < 0.05. Among
overweight women, there was a non-significant effect of
treatment, P ¼ 0.09, but the trend was in the opposite direction. There were no significant effects of race on the percentage of women with excessive weight gains, P > 0.3, nor did
race interact with treatment group or BMI category. Weight
gained before recruitment was strongly related to excessive
weight gain, P < 0.0001 but none of the other demographic
variables predicted excessive weight gain.
Figure 2 Percentage of women whose total weight gain exceeded IOM
recommendations. Overall BMI category by treatment interaction,
P < 0.01; treatment effect among normal-weight women, P < 0.05;
treatment effect among overweight women, P ¼ 0.09.
International Journal of Obesity
Figure 3 Pattern of weight gain at three timepoints: pre-pregnancy
(self-reported), recruitment (mean ¼ 14 weeks), and delivery. The IOM
recommends that overweight women gain 2 lb in the first trimester and
15 – 25 lb (6.8 – 11.4 kg) total, and normal-weight women gain 3.5 lb
(1.6 kg) in the first trimester and 25 – 35 lb (11.4 – 15.9 kg) total.
Continuous analysis of weight gain during pregnancy
Although the intervention affected the percent of normalweight women who exceeded the weight gain goal, Figure 3
shows that the intervention had no significant effects on the
average weight gain from pre-pregnancy to delivery (using
self-reported pre-pregnancy weight) or on weight gain from
recruitment to delivery. On average, normal-weight women
gained 35.0 lb (15.9 kg) from pre-pregnancy to delivery
and overweight women gained 26.5 lb (12.0 kg; P < 0.001).
Weight gains were comparable in black and white women
(31.4 vs 31.0 lb (14.3 vs 14.1 kg)). Weight gain from prepregnancy to delivery was related to pre-pregnancy weight
(P < 0.05), weeks of gestation at delivery (P < 0.05) and the
amount of weight gained before recruitment (P < 0.001), but
was unrelated to any of the baseline demographic variables.
Behavior changes during pregnancy
At recruitment, fat consumption from 13 high-fat foods
averaged 55.9 g per day. All groups decreased their fat consumption from these foods from baseline to 30 weeks, except
normal-weight women in the control condition. There was
no effect of treatment on changes in fat intake from these
foods from recruitment to 30 weeks (P > 0.2), nor was there a
significant treatment by BMI category interaction (P > 0.1).
Twenty women were put on bed rest by their obstetricians; of these women, 11 were put on bed rest prior to the
30-week assessment and were omitted from the exercise
analyses. Mean exercise expenditure at recruitment was
614 kcal=day (2570 kJ=day). There were no differences
between groups in exercise levels at recruitment (P ¼
0.059), but control women tended to be less active than
women in the counseling condition. Changes in exercise
level from recruitment to 30 weeks (P > 0.8) were not related
to treatment condition or BMI. Moreover, collapsing across
Prevention of excessive weight gain in pregnancy
BA Polley et al
1499
treatment groups (n ¼ 110), neither dietary intake nor exercise level was related to the amount of weight gained during
pregnancy nor to whether the women exceeded the IOM
guidelines.
Pregnancy complications and fetal outcomes
Logistic regression was used to examine the effects of
treatment on the percentage of women with low weight
gain, controlling for pre-pregnancy weight and weeks of
gestation at delivery. The percentage of women who gained
less than recommended did not differ significantly for the
intervention vs control group (P > 0.2), nor was the interaction between BMI category and treatment group (P > 0.2)
significant.
Pregnancy outcomes are displayed in Table 4. The number
of obstetric complications was too small for statistical analysis, but does not appear to have been dramatically affected
by the intervention. Similarly, there were no treatment by
BMI category differences in infant birth weight, P > 0.8.
Postpartum weight loss and weight retention
Postpartum weights were available for 74 (67.3%) of the 110
women, who were weighed 8 7.1 (mean s.d.) weeks after
delivery. Postpartum weights were compared with delivery
weights to compute weight loss. The strongest predictor of
postpartum weight loss was the number of weeks postpartum
when weight was measured, P ¼ 0.002; therefore we controlled for weeks postpartum in all analyses of postpartum
weight. As shown in Table 3, the average postpartum weight
loss in each group was about 20 lb (9.1 kg), with no effect of
BMI category or treatment group, nor a significant interaction (all P > 0.4).
Mean postpartum weight retention (postpartum weight
less pre-pregnancy weight) was 8.1 13.1 lb (3.7 5.9 kg),
ranging from 722 to 37 lb (710 to 16.8 kg). Weight retention was strongly correlated with weight gain during pregnancy, r ¼ 0.89, P < 0.001. There was a trend for a significant
interaction between treatment group and BMI category on
weight retention (P ¼ 0.06), with normal-weight women in
the intervention group retaining less than those in the
control group (9.6 vs 13.7 lb; 4.4 kg vs 6.2 kg), and overweight
women in the intervention group retaining more than those
in the control group (7.9 vs 0.66 lb; 3.6 vs 0.3 kg); however,
neither of these differences reached conventional levels of
significance.
Postpartum weight loss and weight retention were also
analyzed by race. There was a significant race effect for
postpartum weight loss (P ¼ 0.01), with black women losing
less weight than white women, with losses of 17.8 and
22.1 lb (8.1 and 10.0 kg), respectively. However, there was
no effect of race on weight retention (postpartum weight less
pre-pregnancy weight), nor did race interact with BMI category or treatment group. Black and white women retained
9.5 and 7.3 lb (4.3 and 3.3 kg), respectively.
Comment
We developed a stepped care, behavioral intervention to
reduce excessive weight gain during pregnancy. The intervention included education and behavioral strategies to
promote healthy, low-fat eating, modest exercise and appropriate weight gain during pregnancy. We found the intervention was effective in reducing the frequency of excessive
weight gain in normal-weight women; whereas 58% of
normal-weight women in the control condition gained
more than 35 lb (15.9 kg) during pregnancy, only 33% of
normal-weight women exceeded this IOM guideline. The
intervention had no significant effect among overweight
women, but the trend was in the opposite direction.
The effectiveness of the intervention among normalweight women appeared to be due primarily to the fact
that fewer women in the intervention group than in the
control group exceeded the expected amount of weight gain
at any point during their pregnancy (63 vs 94%). Such
prevention of excessive weight gain may have resulted
from the basic education about healthy eating and exercise
that was provided to all women in the intervention group
and=or the individual weight graphs they received. There
Table 4 Effect of treatment group and BMI category on pregnancy outcomes
Normal weight
Infant birth weight (g)
Low birth weight (<2500 g), n
Macrosomia (>4000 g), n
Weeks gestation at delivery
Preterm delivery (<36 weeks), n
Cesarean delivery (n)
Pre-eclampsia (n)
Maternal hypertension (n)
Gestational diabetes (n)
Overweight
Control
(n ¼ 31)
Intervention
(n ¼ 30)
Control
(n ¼ 22)
Intervention
(n ¼ 27)
3226.4
3
0
3133.0
4
1
3349.0
2
0
3282.8
1
0
39.5
2
4
0
4
2
39.1
5
2
0
2
0
39.1
3
6
3
4
1
39.4
2
2
2
4
2
International Journal of Obesity
Prevention of excessive weight gain in pregnancy
BA Polley et al
1500
was a trend for the intervention to also help normal-weight
women who exceeded the expected rate of weight gain
‘recover’ and achieve subsequent weight levels within the
IOM guidelines, but this effect was not significant. Thus, the
more intensive stepped-care intervention that occurred
subsequent to excessive weight regain may have been a less
effective component of the intervention than the prevention
component. Since the more intensive stepped-care approach
is the costly aspect of the intervention, further study of
specific components of the intervention is needed.
Whereas the intervention was effective in normal-weight
women, there was a non-significant adverse trend among the
overweight participants. This finding may in part be due to
the low rate of excessive weight gain in the overweight
control group in the present study. While previous studies
suggest that about two-thirds of overweight women may
exceed the IOM guidelines during pregnancy,3,5 only 32%
of the overweight control group in this study exceeded this
level. Of particular note is that only one of nine overweight
black women (11%) in the control condition exceeded the
weight gain goals. Since overweight black women were most
likely to refuse entry into the study, those women who
participated may not have been representative of this population. Alternatively, the lack of a treatment effect among
overweight women may also be related to their longstanding difficulties with weight control, making a minimal
intervention such as the program used here less effective, or
to an unmeasured factor such as dietary restraint, which
could lead to disinhibited overeating if the participants
found the intervention stressful. The differential outcome
for normal and overweight women may also again suggest
that the intervention was more effective for preventing
problems with weight gain than for treating these problems.
Although non-significant, there was a tendency for
normal-weight women in the intervention condition to
gain less than recommended by the IOM. This raises concerns about possible adverse effects of the intervention. The
study was clearly under-powered to examine pregnancy outcomes, but future research should carefully address the safety
of the intervention.
Previous studies suggest that on average women retain
only a small amount of weight post-pregnancy, especially if
potential confounders such as self-report of pre-pregnancy
weight and the effect of aging are taken into consideration.11
In carefully controlled studies, post-pregnancy weight retention in North American women averaged 0.6 – 3.0 kg,12 and
1.4 – 1.6 kg;13 women in the Netherlands retained 1.4 kg14
and women in England retained 70.82 – 0.28 kg11 following
their pregnancy. For a subgroup of women, however, weight
gain during pregnancy may contribute to obesity.11,15 In the
Stockholm Pregnancy and Weight Development Study, 73%
of a sample of severely obese women reported that they
maintained more than 10 kg after pregnancy.15 Over 50%
of Americans are overweight or obese. Most obesity develops
during adulthood and women aged 25 – 44 are at greatest risk
of a major weight gain.16 These large weight gains and the
International Journal of Obesity
resulting development of obesity may relate in part to
pregnancy weight gains. For example, in a study which
followed over 1000 normal-weight women through two
consecutive pregnancies, weight gains above the recommended level during the first pregnancy increased the risk
of being overweight at the time of the second pregnancy
three-fold.17
We found a very strong association (r ¼ 0.89) between
weight gain in pregnancy and postpartum weight retention.
Previous studies have likewise seen such an association with
correlations ranging from 0.36 to 0.84.8,18 – 20 Moreover, in
the present study, the effects of the intervention on postpartum weight retention paralleled the effects of the intervention on excessive weight gain in pregnancy. Specifically,
there was a trend (P ¼ 0.06) for normal-weight women in the
intervention group to retain less than those in the control
group and for overweight women in the intervention group
to retain more weight than those in the control. These
findings support the importance of improving the effectiveness of our intervention for normal-weight women and
developing a better approach for overweight women as a
means of preventing large weight retention postpartum.
Several limitations of the research and its findings should
be noted. Most importantly, the sample size in this study
may have limited our power to detect significant effects of
the intervention. With a sample size of n ¼ 120, we had 0.75
power to detect a difference of 20% between the control and
the intervention group in a one-sided chi-square (a ¼ 0.05),
assuming that 45% of the control group exceeded the IOM
guidelines. Power was limited further by the fact that only
110 of the 120 women completed the study and only 74 were
available for postpartum measurements.
Secondly, this study utilized a lower-income clinic population which faced many barriers to participation in the
program, such as lack of social support (over half were
single), financial hardship (over half were unemployed),
and practical issues such as unreliable transportation for
grocery shopping and access to health care and disruptions
in phone service which interfered with maintaining contact
with the study. Many of the women also faced issues such as
smoking, alcohol and drug use, and coping with an
unplanned pregnancy; this population (and their health
care providers) had more critical concerns than the effect
of excessive weight gain on maternal and infant health.
These women had very limited knowledge about nutrition
and healthful eating and needed intensive education and
counseling; however, their limited resources (eg unreliable
attendance at appointments) made intensive counseling
difficult. Selecting a more adherent population would allow
for a better test of the efficacy of behavioral intervention
during pregnancy.
The intervention used in this study was designed to
minimize risk, and this may have limited its effectiveness.
Participants were initially taught to make healthy, low-fat
choices, but were not prescribed a specific diet or given a
restricted energy intake goal until they had exceeded the
Prevention of excessive weight gain in pregnancy
BA Polley et al
1501
recommended weight at four consecutive clinic visits.
Studies with GDM patients21 – 24 suggest that some energy
restriction can be implemented safely during pregnancy,
although the applicability of these findings to non-diabetic
pregnancies is unclear. This study showed no differences
between conditions in eating or exercise behavior, perhaps
reflecting the limited changes that were obtained, the difficulty of accurately assessing these behaviors,25 and the
limitations of the instruments selected for assessing diet
and activity in the present study. Moreover, data were not
collected on other behaviors that could potentially affect
pregnancy-associated weight changes, including changes in
smoking habits, breast feeding and general lifestyle changes
accompanying pregnancy and motherhood.16
Like most pregnancy studies, we relied on a self-reported
pre-pregnancy weight and for a few cases on self-reported
postpartum weight. Several studies have suggested that such
self-reports are quite accurate, typically underestimating
weight by 1 – 2 kg.17 However, in one study of adolescents,
there was more pronounced under-reporting among the
overweight.27 Thus, self-reported pre-pregnancy weight is a
possible source of error among overweight women, but
should have affected the intervention and control condition
equally.
Weight gain during pregnancy was strongly correlated
with pre-pregnancy weight and with weeks gestation at
delivery. Given this tight biological control, it may be difficult
to influence pregnancy weight gain by lifestyle changes.
Indeed, studies of maternal nutritional supplementation
during pregnancy have found minimal effects on weight
gain, even when supplements of up to 1000 kcal per day are
provided (with notably poor compliance).28,29 This biological
drive, coupled with the numerous competing self-care
demands during pregnancy, may limit the magnitude of
change possible during pregnancy. Thus, a postpartum
approach should also be considered. Lovelady and colleagues30 were able to promote weight loss in newly postpartum
overweight women with a 10 week eating and exercise program, with no detrimental effects on lactation success, and
Leermakers and colleagues31 found that a correspondence
intervention was effective in promoting weight loss in postpartum women. However, during postpartum interventions,
women are busy with infant care and adjustment to motherhood, which may limit the effectiveness of postpartum interventions. Perhaps the most effective approach would be to
begin to educate women during pregnancy about healthy
eating and exercise and attempt to limit the frequency and
extent of excessive pregnancy weight gain, and then continue
such interventions in the postpartum period for those who
have retained excessive amounts of weight.
Our conservative intervention significantly reduced
excessive weight gain among normal-weight women but
not among overweight women. Given that the majority of
childbearing women are normal weight, and that our results
were achieved in a low-SES population with considerable
barriers, this intervention shows promise as a behavioral
method of promoting appropriate weight gain during pregnancy, and helping to prevent the development of long-term
weight problems. Further research is needed to confirm the
efficacy and safety of this type of behavioral intervention for
preventing excessive weight gain in normal-weight women
and to define the key components of the intervention that
are required for its success. In addition, it is important to
develop a more effective way to prevent excessive pregnancy-related weight gain in overweight women.
Acknowledgments
We wish to thank Cathy DeBranski, RN-C for her help in
recruiting and scheduling participants for this study. This
work was funded by a grant from Magee-Womens Health
Foundation; Magee-Womens Research Institute awarded to
Dr Wing.
References
1 Lederman SA. Recent issues related to nutrition during pregnancy. J Am Coll Nutr 1993; 12: 91 – 100.
2 Parker JD, Abrams B. Prenatal weight gain advice: an examination
of the recent prenatal weight gain recommendations of the
Institute of Medicine. Obstet Gynecol 1992; 79: 664 – 669.
3 Keppel KG, Taffel SM. Pregnancy-related weight gain and retention: implications of the 1990 Institute of Medicine guidelines.
Am J Public Health 1993; 83: 1100 – 1103.
4 Scholl TO, Hediger ML, Schall JI, Ances IG, Smith WK. Gestational weight gain, pregnancy outcome, and postpartum weight
retention. Obstet Gynecol 1995; 86: 423 – 427.
5 Carmichael S, Abrams B, Selvin S. The pattern of maternal weight
gain in women with good pregnancy outcomes. Am J Public
Health 1997; 87: 1984 – 1988.
6 Scholl TO, Hediger ML, Ances IG, Belsky DH, Salmon RW. Weight
gain during pregnancy in adolescence: Predictive ability of early
weight gain. Obstet Gynecol 1990; 75: 948 – 953.
7 Muscati, SK, Gray-Donald, K, Koski, KG. Timing of weight gain
during pregnancy: promoting fetal growth and minimizing
maternal weight retention. Int J Obes 1996; 20: 526 – 532.
8 Rossner S, Ohlin A. Pregnancy as a risk factor for obesity: lessons
from the Stockholm pregnancy and weight development study.
Obes Res 1995; 3(Suppl 2): 267s – 275s.
9 Block G, Clifford C, Naughton MD, Henderson M, McAdams M.
A brief dietary screen for high fat intake. J Nutr Educ 1989; 21:
199 – 207.
10 Paffenbarger RS Jr., Hyde RT, Wing AL, Hsieh C-C. Physical
activity, all-cause mortality, and longevity of college alumni.
New Engl J Med 1986; 314: 605 – 613.
11 Harris, HE, Ellison, GTH, Holliday, M, Lucassen, E. The impact
of pregnancy on the long-term weight gain of primiparous
women in England. Int J Obes Relat Metab Disord 1997; 21:
747 – 755.
12 Smith, DE, Lewis, CE, Caveny, JL, Perkins, LL, Burke, GL, Bild, DE.
Longitudinal changes in adiposity associated with pregnancy.
JAMA 1994; 271: 1747 – 1751.
13 Williamson, DF, Madans, J, Pamuck, E, Flegal, KMk Kendrick, JS,
Serdula, MK. 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: 561 – 569.
14 Rookus, MA, Rokebrand, P, Burema, J, Deurenberg, P. The effect of
pregnancy on the body mass index 9 months post-partum in 49
women. Int J Obes 1987; 11: 609 – 618.
15 Rossner, S. Pregnancy, weight cycling and weight gain. Int J Obes
1992; 16: 145 – 147.
International Journal of Obesity
Prevention of excessive weight gain in pregnancy
BA Polley et al
1502
16 Williamson, DF, Kahn, HS, Remington, PL, Anda, RF. The 10-year
incidence of overweight and major weight gain in US adults. Arch
Intern Med 1990; 150: 665 – 672.
17 Gunderson, EP, Abrams, B, Selvin, S. The relative importance of
gestational gain and maternal characteristics associated with the
risk of becoming overweight after pregnancy. Int J Obes Relat
Metab Disord 2000; 24: 1660 – 1668.
18 Hunt SC, Daines MM, Adams TD, Heath EM, Williams RR.
Pregnancy weight retention in morbid obesity. Obes Res 1995;
3: 121 – 130.
19 Schauberger CW, Rooney BL, Brimer LM. Factors that
influence weight loss in the puerperium. Obstet Gynecol 1992;
79: 424 – 429.
20 Parham ES, Astrom MF, King SH. The association of pregnancy
weight gain with the mother’s postpartum weight. J Am Diet Assoc
1990; 90: 550 – 554.
21 Algert S, Shragg P, Hollingsworth DR. Moderate caloric restriction
in obese women with gestational diabetes. Obstet Gynecol 1985;
65: 487 – 491.
22 Coetzee EJ, Jackson WPU, Berman PA. Ketonuria in pregnancy —
with special reference to calorie-restricted food intake in obese
diabetics. Diabetes 1980; 29: 177 – 181.
23 Maresh M, Gillmer MDG, Beard RW, Alderson CS, Bloxham BA,
Elkeles RS. The effect of diet and insulin on metabolic profiles of
women with gestational diabetes mellitus. Diabetes 1985;
34(Suppl 2): 88 – 93.
International Journal of Obesity
24 Dornhurst A, Nicholls SD, Probst F, et al. Calorie restriction for
treatment of gestational diabetes. Diabetes 1991; 40(Suppl 2):
161 – 164.
25 Lichtman SW, Pisarka, K et al. Discrepancy between self-reported
and actual caloric intake and exercise in obese subjects. New Engl J
Med 1992; 327: 1893 – 1898.
26 Harris HE, Ellison GTH, Clement S. Do the psychosocial and
behavioral changes that accompany motherhood influence the
impact of pregnancy on long-term weight gain. Obstet Gynecol
1999; 20: 65 – 79.
27 Stevens-Simon C, Roghmann KJ, McAnarney ER. Relationship of
self-reported pre-pregnant weight and weight gain during pregnancy to maternal body habitus and age. J Am Diet Assoc 1992;
92: 85 – 87.
28 Adair LS, Pollitt E, Mueller WH. The Bacon Chow study: Effect of
nutritional supplementation on maternal weight and skinfold
thicknesses during pregnancy and lactation. Br J Nutr 1984; 51:
357 – 369.
29 Rush D, Sloan NL, Leighton J et al. Longitudinal study of pregnant women. Am J Clin Nutr 1988; 48: 439 – 483.
30 Lovelady CA, Garner KE, Moreno KL, Williams JP. The effect of
weight loss in overweight, lactating women on the growth of
their infants. New Engl J Med 2000; 342: 449 – 453.
31 Leermakers EA, Anglin K, Wing RR. Reducing postpartum weight
retention through a correspondence intervention. Int J Obes Relat
Metab Disord 1998; 22: 1103 – 1109.