The Effect of an Exercise Intervention on Gestational Weight Gain: The Behaviors Affecting Baby and You (B.A.B.Y.) Study: A Randomized Controlled Trial

Abstract

Purpose:

To examine the effect of a prenatal exercise intervention on gestational weight gain (GWG) and to update meta-analyses.

Design:

Randomized controlled trial and meta-analysis.

Setting:

Obstetrical practices in a Western Massachusetts hospital.

Patients:

We analyzed 241 ethnically diverse pregnant participants at high risk for gestational diabetes in the Behaviors Affecting Baby and You (B.A.B.Y.) study. A total of 118 participants were randomized to an exercise intervention group and 123 to a comparison health and wellness intervention group.

Intervention:

A 12-week individually tailored, motivationally matched program designed to increase the compliance with guidelines for exercise during pregnancy (30 min/day).

Measures:

The GWG and compliance with 2009 Institute of Medicine (IOM) guidelines for GWG abstracted from medical records.

Analysis:

Unadjusted logistic regression, intent-to-treat. Results were added to the existing meta-analyses using a random effects model.

Results:

Women randomized to the exercise group had a lower mean GWG than the comparison group (−0.97 kg, P value = .39) and were less likely to exceed IOM guidelines (odds ratio = 0.69, 95% confidence interval [CI] 0.34-1.40), but results were not statistically significant. Meta-analyses yielded a −0.63 kg (95% CI −1.17 to −0.08, P = .02) reduction in GWG and a 20% reduction in odds of exceeding IOM guidelines (95% CI 0.73 to 0.89) for the exercise intervention.

Conclusion:

Findings from this randomized trial among ethnically diverse women contribute to the results of meta-analyses supporting exercise as a means of attenuating GWG.

Keywords

exercise gestational weight gain Latina pregnancy

Purpose

Excessive gestational weight gain (GWG) is common during pregnancy, with 48% of women in the United States exceeding Institute of Medicine (IOM) guidelines.¹ Compared to previous decades, women in the United States currently enter pregnancy at higher weights² and are more likely to gain excess weight during pregnancy.^1,3,4 Excess GWG has been associated with pregnancy complications, including gestational diabetes,^5,6 cesarean delivery,^7,8 large-for-gestational-age infants,^9,10 and postpartum weight retention.^11,12 Just as importantly, excessive GWG has been associated with a long-term risk of overweight and obesity for both mother and baby.^13
-15 Ethnic minorities such as Hispanics are disproportionately affected by overweight and obesity,^16
-18 and almost half begin their pregnancies overweight or obese,^17,19

-22 with rates of excessive GWG increasing over time.^3,4

Identifying interventions that prevent excess weight gain during pregnancy among ethnically diverse populations is therefore critical. However, it remains unclear as to which intervention types yield the best outcomes and whether interventions work equally for all risk groups.²³ Few individual trials of exercise interventions during pregnancy have been adequately powered to detect differences in GWG, and, in light of this concern, recent meta-analyses, including a Cochrane Review, have been conducted.^23
-25 These meta-analyses included 22 studies and found that women assigned to prenatal exercise interventions experienced small reductions in mean GWG ranging from −0.61 (95% confidence interval [CI] −1.17 to −0.06; P = .03)²⁴ to −1.00 kg (95%CI −2.10 to 0.01; P = .051)²³ versus comparison groups. The Cochrane Review repeated their meta-analysis among high-risk pregnant women (defined as overweight/obese or at risk for gestational diabetes mellitus [GDM]) and found a smaller −0.34 kg reduction in GWG (95% CI −1.15 to 0.47; P = .41).²³ Heterogeneity ranged from 25%²⁴ to 60.0%,²³ making it difficult to determine the intervention types that are most effective in making clinically meaningful changes in GWG as well as the ideal characteristics of such interventions including frequency, duration, and intensity. In addition, the majority of studies included in the meta-analysis utilized group–exercise intervention models and/or required regular monitoring of activity by study personnel, which may not be feasible in regular prenatal care settings. Finally, no prior interventions included a significant proportion of racially/ethnically diverse women.

Therefore, we conducted a randomized trial of an individually tailored, motivationally matched exercise intervention among an ethnically diverse group of pregnant women at high risk of GDM. Our goals were 2-fold: (1) to evaluate the impact of a prenatal exercise program on GWG and meeting IOM GWG guidelines and (2) to update 2 recent meta-analyses with our findings.

Methods

Design

The Behaviors Affecting Baby and You (B.A.B.Y.) study was a randomized trial conducted between 2007 and 2012 among a diverse group of pregnant women receiving prenatal care at the ambulatory obstetrical practices of Baystate Medical Center (Figure 1).²⁶ The overall goal of the B.A.B.Y. study was to assess the impact of the intervention on risk of GDM, and therefore eligibility was restricted to women at high risk of GDM which is defined as (1) personal history of GDM in a previous pregnancy or (2) body mass index (BMI) >25 kg/m² and a family history of type 2 diabetes. All women signed a written informed consent and the study was approved by the institutional review boards of the University of Massachusetts Amherst and Baystate Medical Center.

Figure 1.

Study design—B.A.B.Y. study, 2007-2012. B.A.B.Y. indicates The Behaviors Affecting Baby and You.

Sample

Women were excluded from the study if they engaged in more than 30 minutes of moderate or vigorous activity on 3 or more days per week, were younger than 18 or older than 40 years of age, were at more than 20 weeks’ gestation, were unable to read English at a sixth-grade level, did not plan to deliver at Baystate Medical Center, had a nonsingleton pregnancy, had contraindications to engaging in moderate exercise, were currently taking medications that adversely influenced glucose tolerance or had a personal history of diabetes, hypertension, heart disease, or chronic renal disease. For the purposes of the current analysis, we also excluded women who developed medical contraindications to exercise, who experienced a miscarriage or termination, or who were missing information on GWG.

Intervention

Eligible participants were randomized in blocks of 8 based on age, prepregnancy BMI, and ethnicity into either the exercise intervention group or a comparison health and wellness intervention group. Women were not blinded to their assigned intervention group. The interventions were 12-week programs designed to end at or near the time of routine GDM screen at 24 to 28 weeks’ gestation. Contact time was standardized across groups, with both receiving one face-to-face visit at baseline, and weekly and biweekly telephone booster calls and mailed informational materials. All prenatal care patients at Baystate also received counseling for appropriate nutrition and weight gain in pregnancy.²⁷

The goal of the exercise intervention was to encourage participants to meet the American Congress of Obstetricians and Gynecologists (ACOG) criteria for exercise during pregnancy (30 min/d on most days of the week).²⁸ The exercise intervention was tailored to each participant, taking into account her baseline exercise, her readiness to change, and her lifestyle and preferences for exercise. Women were encouraged to increase their exercise 10% each week from their baseline level and given flexibility in choosing the timing and type of exercise, with an emphasis on increased walking. Participant’s stage of change along with their self-efficacy, barriers and benefits of exercise, and processes of change for exercise were used to provide stage-matched motivational targeted content. Weekly and biweekly booster telephone calls provided motivationally based individualized feedback.

Women in the comparison group received a 12-week health and wellness intervention focused on general issues related to health and wellness during pregnancy. Mailings of ACOG informational brochures on healthy pregnancy topics represented high-quality, low-cost self-help material generally available to the public.

Measures

Total gestational weight gain (in kilograms) was abstracted from medical records after delivery by medical record abstractors blinded to the study arm and was calculated as the difference between measured maternal weight at a prenatal care appointment closest to delivery (mean [SD] = 39.0 (1.1) weeks’ gestation) and prepregnancy weight. Prepregnancy weight was either self-reported to the interviewer at the time of recruitment (77.6%) or self-reported to the practitioner at the time of the first prenatal care visit and recorded in the medical record (11.6%). If prepregnancy weight was not available from either of these sources, it was based upon measured weight at the first prenatal care visit (10.8%). The validity of self-reported prepregnancy weight has been shown to be high, especially if collected early in pregnancy.^29
-31 A recent validation study found a strong correlation (r = .95, P = .0001) between self-reported prepregnancy weight and physician measured weight from the year before pregnancy with a mean discrepancy of 0.5 ± 3.0 kg and no significant (P = .64) differences between normal weight and overweight/obese participants.³²

The rate of GWG during the second and third trimesters was calculated using the ratio of the change in weight from weight measured at the prenatal visit closest to 13 weeks’ gestation (mean [SD] = 13.2 (1.3) weeks’ gestation) to delivery weight and the difference in gestational age (in weeks) between the 2 measures.³³ Meeting IOM guidelines for GWG and rate of GWG was defined as a 3-level outcome variable: inadequate, adequate (referent group), or excessive weight gain during pregnancy.³³

Sociodemographic factors, including ethnicity, education, annual household income, marital status, and health behaviors, including alcohol and tobacco use both pre-pregnancy and during early pregnancy, were assessed through a structured questionnaire given at recruitment. Information on other types (occupational, household, and transportation) and intensities (vigorous, moderate, light and sedentary) of physical activity were also assessed using the 33-item Pregnancy Physical Activity Questionnaire administered at baseline and at the end of the intervention.³⁴

Analysis

All analyses were carried out as intent to treat. Differences in sociodemographic, medical history, and behavioral characteristics were compared across study groups using chi-square or Fisher exact tests. Linear regression was used to model the mean difference (MD) in GWG and rate of GWG by study group. Unadjusted and adjusted estimates, standard errors, and P values were computed. Multinomial logistic regression was used to model the odds of excessive and inadequate total versus adequate GWG by study group. Multinomial logistic regression was also used to model the odds of excessive and inadequate rate of GWG versus adequate rate of GWG by study group. Unadjusted and adjusted odds ratios (ORs), 95% CIs, and P values were computed. All analyses were conducted in SAS version 9.4 (SAS Institute Inc, Cary, North Carolina).

We then conducted 2 sensitivity analyses to take into account adherence to the study protocol. First, we defined adherence as dose of intervention received in weeks. Second, we defined adherence as meeting the ACOG guidelines for physical activity by the end of the intervention and compared those participants in the exercise group who met the ACOG guidelines to the entire health and wellness group.

Because the primary goal of the B.A.B.Y. study was to evaluate the impact of the intervention on GDM risk, the intervention was powered for outcomes of glucose abnormalities. However, based on our final analytic sample size of 246, we had 80% power to detect a MD of 2.5 kg in GWG between study groups and 80% power to detect an OR of 0.46 or more for exceeding IOM guidelines.³⁵

We updated the recent meta-analyses of physical activity interventions on GWG by Streuling et al and by the Cochrane Review with our findings.^23,24 Specifically, we used a random effects model in the metafor package in R.^36,37 The random effects model for the MD in weight change during pregnancy incorporate the mean GWG, its standard deviation, and the group size for both the exercise intervention and comparison groups from each study to generate a summary estimate of effect. The random effects model for the risk of excessive GWG incorporates the proportion exceeding IOM guidelines and group size for both the exercise intervention and comparison groups from each study. Heterogeneity was tested with Cochrane’s Q and Higgins’ I ². We did not update a recent meta-analysis on dietary and lifestyle interventions by Thangaratinam et al²⁵ as this article did not provide sufficient data on the individual trials included in analysis.

Results

Of the 488 participants initially enrolled in the B.A.B.Y. study, 290 met study inclusion criteria and were randomized into the exercise or health and wellness group at a mean of 11.8 (3.4) weeks’ gestation (Figure 2). Retention to delivery was 83.1% and did not differ between the study groups (P = .79). Specifically, after randomization, participants were additionally excluded from the final analysis if they had medical contraindications to exercise (n = 4), experienced a miscarriage/termination of pregnancy (n = 3), or were missing data on delivery weight (n = 42), leaving a final study population of 241 (118 [49.0%] in the exercise group and 123 [51.0%] in the health and wellness group).

Figure 2.

Participant flowchart—B.A.B.Y. study, 2007-2012. B.A.B.Y. indicates The Behaviors Affecting Baby and You.

The majority of study participants were young (49.4% less than 25 years) and Hispanic (61.4%) and 54.0% had household incomes less than $30 000/year (Table 1). Most participants reported living with a partner (58.9%) and one-quarter were married (25.7%). The majority of participants were parous (72.6%) and overweight (34.9%) or obese (62.2%).

Table 1.

Distribution of Covariates by Intervention Group: B.A.B.Y. Study 2007-2012.^a

	Total		Exercise Intervention (n = 118)		Health & Wellness Intervention (n = 123)
	n	%	n	%	n	%	P value*
Demographic characteristics
Age (years)
18-19	25	10.4	15	12.7	10	8.1	.33
20-24	94	39.0	48	40.7	46	37.4
25-29	58	24.1	23	19.5	35	28.5
30-40	64	26.6	32	27.1	32	26.0
Ethnicity
Hispanic	148	61.4	68	57.6	80	65.0	.24
Non-Hispanic	93	38.6	50	42.4	43	35.0
Education
Less than high school	56	25.3	25	22.5	31	28.2	.07
High school graduate or GED	76	34.4	33	29.7	43	39.1
Post high school	89	40.3	53	47.8	36	32.7
Marital status
Single/separated/divorced/ widowed	177	74.1	85	72.0	92	76.0	.48
Married	62	25.9	33	28.0	29	24.0
Live with partner
Yes	142	64.0	72	64.9	70	63.1	.78
No	80	36.0	39	35.1	41	36.9
Adults in household
1	58	25.2	25	21.7	33	28.7	.46
2	115	50.0	61	53.0	54	47.0
3 or more	57	24.8	29	25.2	28	24.4
Children in household
0	53	22.7	30	25.6	23	19.7	.46
1	88	37.6	45	38.5	43	36.8
2	64	27.4	27	23.1	37	31.6
3 or more	29	12.4	15	12.8	14	12.0
Health status
Parity
Nulliparous	66	27.4	40	33.9	26	21.1	.03
Parous	175	72.6	78	66.1	97	78.9
Prepregnancy BMI (kg/m²)
Less than 25	6	2.5	3	2.5	3	2.4	.44
25 to less than 30	79	32.8	34	28.8	45	36.6
30 or greater	156	64.7	81	68.6	75	61.0
Personal history of GDM
Yes	23	10.4	6	5.6	17	14.9	.02
No	199	89.6	102	94.4	97	85.1
Family history of diabetes mellitus
Yes	224	98.7	110	99.1	114	98.3	1.00
No	3	1.3	1	0.9	2	1.7
Smoking during pregnancy
Yes	32	14.9	16	15.2	16	14.6	.89
No	183	85.1	89	84.8	94	85.5
Alcohol use during pregnancy
Yes	3	1.4	3	2.8	0	0	.12
No	216	98.6	106	97.3	110	100.0
GDM in current pregnancy
Yes	30	13.5	12	10.9	18	15.9	.27
No	193	86.6	98	89.1	95	84.1

Abbreviations: B.A.B.Y., The Behaviors Affecting Baby and You; GDM, gestational diabetes mellitus.

^aN = 241. Numbers may not total to N = 241 due to missing data.

^bPearson chi-square test or Fisher exact test (if cell sizes are less than 5). Excludes missing values and “don’t know”/“refused.”

The study groups did not differ according to the sociodemographic and behavioral factors and mean (SD) gestational age at delivery (39.2 [1.8] weeks) did not differ between groups (P = .63). However, participants in the exercise group were less likely to be parous (66.1% vs 78.9%; P = .03) and less likely to have a history of GDM in a previous pregnancy (5.1% vs 13.8%; P = .02) as compared to participants in the health and wellness group (Table 1).

Mean GWG (SD) was 12.9 (8.3) kg in the exercise group and 13.8 (9.0) kg in the health and wellness group (P = .39; Table 2). In the exercise group, 64.4% had excessive and 17.0% inadequate GWG according to IOM guidelines, as compared to 68.3% and 17.9%, respectively, in the health and wellness group (P = .59; Table 2).

Table 2.

Distribution of Measures of Gestational Weight Gain and Meeting Institute of Medicine (IOM) Guidelines by Intervention Group: B.A.B.Y. Study 2007-2012.^a

	Total		Exercise Intervention (n = 118)		Health and Wellness Intervention (n = 123)
	Mean	SD	Mean	SD	Mean	SD
Measures of gestational weight gain
Total pregnancy gestational weight gain (kg)	13.4	8.7	12.9	8.3	13.8	9.0
Rate of weight gain (kg/wk)	0.43	0.27	0.44	0.29	0.41	0.25
Meeting IOM guidelines	n	%	n	%	n	%
Gestational weight gain
Not achieving IOM guidelines	41	17.0	20	17.0	21	17.1
Meeting IOM guidelines	39	16.2	22	18.6	17	13.8
Exceeding IOM guidelines	161	66.8	76	64.4	85	69.1
Rate of gestational weight gain
Not achieving IOM guidelines	42	19.8	23	21.3	19	18.3
Meeting IOM guidelines	21	9.9	10	9.3	11	10.6
Exceeding IOM guidelines	149	70.3	75	69.4	74	71.2

Abbreviations: B.A.B.Y., The Behaviors Affecting Baby and You; SD, standard deviation.

^aN = 241.

The mean rate of GWG during the second and third trimester was 0.44 (0.29) kg/wk in the exercise group and 0.41 (0.25) kg/wk in the health and wellness group (P = 0.55). Rates of excessive and inadequate GWG according to IOM guidelines were 69.4% and 21.3% in the exercise group as compared to 70.2% and 18.3% in the health and wellness group (P = .84; Table 2).

We then evaluated unadjusted and adjusted MDs in GWG and rate of GWG by study group (Table 3). The exercise arm had, on average, a −0.97 kg reduction in GWG (95% CI −3.17 to 1.23; P = .39) as compared to the health and wellness group. Findings were slightly strengthened but remained nonstatistically significant after adjusting for parity and personal history of GDM in a previous pregnancy (−1.67 kg, 95% CI −3.91 to 0.56 kg; P = .14). Similarly, although rate of weight gain appeared slightly lower in the exercise group in adjusted analyses (β= −0.01 kg/wk, SE 0.04, 95% CI −0.09 to 0.06; P = .74), this difference was also not statistically significant (Table 3). Adjusting for incident GDM during the current pregnancy also did not change study findings.

Table 3.

Impact of the Physical Activity Intervention on Measures of Gestational Weight Gain and Meeting Institute of Medicine (IOM) Guidelines: Intent-to-Treat Analysis—BABY Study 2007-2012.^a

	Unadjusted^b				Adjusted^b
	β	SE	95% CI	P value	β	SE	95% CI	P value
Measures of gestational weight gain
Total pregnancy gestational weight gain (kg)	−0.97	1.12	(−3.17-1.23)	.39	−1.67	1.13	(−3.91-0.56)	.14
Rate of weight gain (kg/wk)	0.02	0.04	(−0.05-0.10)	.55	−0.01	0.04	(−0.09-0.06)	.74
Meeting IOM guidelines	OR		95% CI	P value	OR		95% CI	P−value
Gestational weight gain
Not achieving IOM guidelines	0.74		(0.31-1.78)	.50	0.94		(0.37-2.42)	.90
Meeting IOM guidelines	Referent				referent
Exceeding IOM guidelines	0.69		(0.34-1.40)	.30	0.74		(0.35-1.57)	.43
Rate of gestational weight gain
Not achieving IOM guidelines	1.33		(0.47-3.81)	.59	1.59		(0.52-4.89)	.42
Meeting IOM guidelines	Referent				referent
Exceeding IOM guidelines	1.12		(0.45-2.78)	.82	1.02		(0.38-2.71)	.97

Abbreviations: B.A.B.Y., The Behaviors Affecting Baby and You; CI, confidence interval; GDM, gestational diabetes mellitus; OR, odds ratio; SE, standard error.

^aN = 241.

^bLinear regression to calculate mean differences for continuous outcomes and multinomial logistic regression for categorical outcomes; multivariable models adjusted for parity and personal history of GDM.

In terms of meeting IOM guidelines for GWG, the exercise arm had a 30% decreased odds of exceeding IOM guidelines (OR 0.69, 95% CI 0.34-1.40) as compared to the health and wellness group, but this was not statistically significant (Table 3). Findings were similar after adjusting for parity and personal history of GDM. In terms of rate of weight gain, findings were also not statistically significant for odds of excessive rate of weight gain in the exercise versus the comparison group (OR 1.12, 95% CI 0.45-2.78; Table 3).

We then updated the meta-analysis of physical activity intervention trials and GWG by Streuling et al by including a total of 12 studies and 906 participants.²⁴ Updated findings (MD −0.63 kg, 95% CI −1.17 to −0.08; P = .03) were similar to the original findings (MD −0.61 kg, 95% CI −1.17 to −0.06; P = .03), and we did not observe significant heterogeneity (I ² = 25.90%, Q = 14.8, P = .25). We also updated the Cochrane Review meta-analysis of 4 studies of 476 high-risk participants which led to slightly stronger findings for GWG (original MD −0.32 kg, 95% CI −1.15 to 0.50; P = .44; updated MD −0.40 kg, 95% CI −1.18 to 0.37; P = .31),²³ again without significant heterogeneity (I ² = 0%, Q = 1.77, P = .78). The Cochrane Review additionally evaluated odds of excessive GWG among 6 studies that included all levels of baseline risk and found that those assigned to exercise interventions had an OR of 0.79 (95% CI 0.70-0.89) for excessive GWG as compared to those assigned to comparison groups.²³ Updating these results with our findings resulted in a comparable OR of 0.80 (95% CI 0.73-0.89) for excessive GWG.

We then conducted a sensitivity analysis according to the dose of intervention. The mean (SD) intervention dose was 9.9 (4.7) weeks (IQR 6.4-13.1). The odds of GDM did not differ according to the dose of intervention; women who received 10 or more weeks of the intervention (72%) had an OR of GDM of 0.62 (95% CI 0.23-1.68) as compared to women who received less than 10 weeks of intervention (28%; OR = 0.60; 95% CI 0.18-2.07).

We then conducted a sensitivity analysis in which we compared those participants in the exercise group who met the ACOG guidelines for physical activity to the entire health and wellness group. A total of 63.6% of participants in the exercise group met this criteria. In comparison to the entire health and wellness group, women in the exercise group who met the ACOG guidelines for physical activity had a lower mean GWG (−0.94 kg; 95% CI −3.5 to 1.6) and a lower odds of excessive GWG (OR = 0.69; 95% CI 0.31-1.15).

Discussion

In this randomized trial of an ethnically diverse group of pregnant women at high-risk for GDM, we found that an individually tailored, motivationally matched exercise intervention resulted in an almost 1 kg lower mean GWG and a 30% decreased odds of exceeding IOM guidelines versus the comparison group, but neither finding was statistically significant. However, the magnitude of these results was consistent with previous meta-analyses which found a statistically significant 0.63 kg reduction in GWG and a 20% reduction in odds of exceeding GWG guidelines. Therefore, our findings add to the evidence base supporting the effectiveness of pregnancy physical activity interventions in reducing gestational weight gain.

Given the high prevalence of US pregnant women exceeding IOM guidelines for GWG and the growing rates of obesity, these finding are important on a population-wide level and support exercise as a means of attenuating GWG.

Interpretation

In the study most similar to our design, Renault et al randomized 425 obese (≥30 kg/m²) participants in Copenhagen to a home-based intervention or to standard care.³⁸ The home-based intervention consisted of walking intervention with a goal of 11,000 steps per day, with pedometer data regularly monitored by study staff to gauge adherence. Median GWG was significantly lower in the intervention group as compared to standard care (9.4 vs 10.9 kg, respectively; P = .042).³⁸ Similarly, we observed an MD of 12.9 kg versus 13.8 kg (P = .39) between the exercise versus the comparison group in our smaller study of 241 participants. The Renault study provides support for the potential efficacy of home-based walking interventions in lowering GWG among women at high risk for excessive GWG and suggests that a larger sample size and regular monitoring of activity levels to help promote adherence may be needed to ensure appropriate power to detect changes in mean GWG.

Strengths and Limitations

Our study had several strengths compared to prior trials including a high reach, low-cost strategy that is readily translatable to clinical practice as well as the inclusion of a significant proportion of racially/ethnically diverse women. Our study also faced several limitations. Adherence to intervention content is a potential concern for home-based interventions as compared to supervised group interventions. However, in recently published article evaluating the effect of the exercise intervention on change in activity in the B.A.B.Y. study showed that the exercise intervention group increased their sports/exercise activity from a mean (SD) of 7.9 (11.2) to 13.1 (11.4) metabolic equivalent of task (MET) hours/week, while the health and wellness group increased their sports/exercise from 6.7 (7.8) to 7.0 (9.1) MET hours/week.³⁹ The difference in change between groups (5.3 vs 0.3) was statistically significant (P = .002). In addition, the exercise group was more likely to achieve the ACOG guidelines for physical activity as compared to the health and wellness group (OR = 2.12; 95% CI 1.45-3.10).

While all participants in the B.A.B.Y. study received dietary counseling as a part of regular care, the study intervention did not include additional dietary counseling. A combined exercise and dietary intervention may have resulted in stronger findings and therefore may be an important direction for future research. Additionally, our study relied on self-report of prepregnancy weight, which may have introduced misclassification of the assessment of GWG. However, self-reported prepregnancy weight has been highly correlated with physician-measured weight in the year prior to pregnancy.³² Any misreporting of prepregnancy weight is expected to be independent of assignment to study group and to bias results toward the null.

Because parity and history of GDM were not randomly distributed between the study arms, we adjusted for these variables in our final models to avoid potential confounding. However, there were no differences in the statistical significance of study findings regardless of adjustment for parity and history of GDM and confidence intervals largely overlapped. These observations suggest that parity and history of GDM were not a significant source of bias. Finally, study findings may be generalized to women without medical contraindications to exercise in pregnancy who are interested in and are able to participate in a prenatal exercise intervention study.

Because the studies included in the meta-analysis differed widely in their design and methods, the meta-analysis was unable to evaluate the aspects of the intervention methods that had the greatest impact on GWG. For example, the individual studies differed in utilization of group versus individual exercise, in types of activity, in volume of activity recommended per week, in gestational age at initiation, and in the duration of the intervention during pregnancy. However, the fact that aggregate findings in meta-analyses translate to small, but significant, reductions in GWG and odds of excessive gestational weight gain, while few individual studies report significant findings highlight the low statistical power of most pregnancy exercise interventions. We had 80% power to observe a statistically significant reduction in GWG of −2.5 kg but observed an effect size of −0.97 kg.

Conclusions

In summary, we found that pregnant women participating in an individually tailored motivationally matched exercise intervention during pregnancy had a modest but statistically nonsignificant lower mean GWG and lower odds of exceeding IOM guidelines as compared to women participating in a comparison health and wellness intervention. However, with this study’s contribution, meta-analyses show that prenatal physical activity interventions can lead to a small, but statistically significant reduction in GWG and odds of exceeding GWG guidelines. Given the high proportion of pregnant women exceeding IOM guidelines for GWG and the growing rates of obesity, these findings are important on a population-wide level and support exercise as a means of attenuating GWG.

SO WHAT? Implications for Health Promotion Practitioners and Researchers

What is already known on this topic?

Participating in an exercise intervention during pregnancy may be associated with a lower risk of excessive gestational weight gain (GWG). However, the majority of past trials utilized group exercise models difficult to implement in clinical practice and were conducted among predominantly non-Hispanic white women.

What does this article add?

Women randomized to a 12-week high reach, low-cost intervention had lower GWG and were less likely to exceed Institute of Medicine (IOM) guidelines than a comparison group, although results were not statistically significant. Findings contribute to meta-analyses supporting exercise as a means of attenuating GWG.

What are the implications for health promotion practice or research?

Given the high prevalence of pregnant women exceeding IOM guidelines and the growing rates of obesity, findings are important on a population-wide level and support exercise as a means of attenuating GWG.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by NIH/NIDDK grant R01DK074876.

References

Dalenius

Brindley

Smith

Reinold

Grummer-Strawn

. Pregnancy Nutrition Surveillance 2010 Report. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2012.

Ogden

Carroll

Kit

Flegal

. Prevalence of childhood and adult obesity in the united states, 2011-2012. JAMA. 2014;311(8):806–814.

Helms

Coulson

Galvin

. Trends in weight gain during pregnancy: a population study across 16 years in north carolina. Am J Obstet Gynecol. 2006;194(5): e32–e34.

Schieve

Cogswell

Scanlon

. Trends in pregnancy weight gain within and outside ranges recommended by the institute of medicine in a WIC population. Matern Child Health J. 1998;2(2):111–116.

Carreno

Clifton

Hauth

; Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network. Excessive early gestational weight gain and risk of gestational diabetes mellitus in nulliparous women. Obstet Gynecol. 2012;119(6):1227–1233.

Hedderson

Gunderson

Ferrara

. Gestational weight gain and risk of gestational diabetes mellitus. Obstet Gynecol. 2010;115(3):597–604.

Chung

Taylor

Thompson

; SCOPE Consortium. Gestational weight gain and adverse pregnancy outcomes in a nulliparous cohort. Eur J Obstet Gynecol Reprod Biol. 2013;167(2):149–153.

Mochhoury

Razine

Kasouati

Kabiri

Barkat

. Body mass index, gestational weight gain, and obstetric complications in moroccan population. J Pregnancy. 2013;2013:379461.

Cedergren

. Effects of gestational weight gain and body mass index on obstetric outcome in sweden. Int J Gynaecol Obstet. 2006;93(3):269–274.

10.

Chen

Shao

. Prepregnancy body mass index, gestational weight gain, and pregnancy outcomes in china. Int J Gynaecol Obstet. 2010;109(1):41–44.

11.

Martin

Hure

Macdonald-Wicks

Smith

Collins

. Predictors of post-partum weight retention in a prospective longitudinal study. Matern Child Nutr. 2014;10(4):496–509.

12.

Rode

Kjaergaard

Ottesen

Damm

Hegaard

. Association between gestational weight gain according to body mass index and postpartum weight in a large cohort of danish women. Matern Child Health J. 2012;16(2):406–413.

13.

McClure

Catov

Ness

Bodnar

. Associations between gestational weight gain and BMI, abdominal adiposity, and traditional measures of cardiometabolic risk in mothers 8 y postpartum. Am J Clin Nutr. 2013;98(5):1218–1225.

14.

Fraser

Tilling

Macdonald-Wallis

. Associations of gestational weight gain with maternal body mass index, waist circumference, and blood pressure measured 16 y after pregnancy: the Avon Longitudinal Study of Parents and Children (ALSPAC). Am J Clin Nutr. 2011;93(6):1285–1292.

15.

Mamun

Mannan

Doi

. Gestational weight gain in relation to offspring obesity over the life course: a systematic review and bias-adjusted meta-analysis. Obes Rev. 2014;15(4):338–347.

16.

Pan

Freedman

Gillespie

Park

Sherry

. Incidences of obesity and extreme obesity among US adults: findings from the 2009 behavioral risk factor surveillance system. Popul Health Metr. 2011;9(1):56. doi:10.1186/1478-7954-9-56.

17.

Hinkle

Sharma

Kim

. Prepregnancy obesity trends among low-income women, united states, 1999-2008. Matern Child Health J. 2012;16(7):1339–1348.

18.

Zambrana

Logie

. Latino child health: need for inclusion in the US national discourse. Am J Public Health. 2000;90(12):1827–1833.

19.

Taveras

Gillman

Kleinman

Rich-Edwards

Rifas-Shiman

. Racial/ethnic differences in early-life risk factors for childhood obesity. Pediatrics. 2010;125(4):686–695.

20.

Chasan-Taber

Schmidt

Pekow

Sternfeld

Solomon

Markenson

. Predictors of excessive and inadequate gestational weight gain in hispanic women. Obesity (Silver Spring). 2008;16(7):1657–1666.

21.

Sam

Berry

. Acculturation when individuals and groups of different cultural backgrounds meet. Perspect Psychol Sci. 2010;5(4):472–481.

22.

Kieffer

Tabaei

Carman

Nolan

Guzman

Herman

. The influence of maternal weight and glucose tolerance on infant birthweight in latino mother-infant pairs. Am J Public Health. 2006;96(12):2201–2208.

23.

Muktabhant

Lawrie

Lumbiganon

Laopaiboon

. Diet or exercise, or both, for preventing excessive weight gain in pregnancy. Cochrane Database Syst Rev. 2015;(6): CD007145.

24.

Streuling

Beyerlein

Rosenfeld

Hofmann

Schulz

von Kries

. Physical activity and gestational weight gain: a meta-analysis of intervention trials. BJOG. 2011;118(3):278–284.

25.

Thangaratinam

Rogozinska

Jolly

. Effects of interventions in pregnancy on maternal weight and obstetric outcomes: meta-analysis of randomised evidence. BMJ. 2012;344: e2088.

26.

Chasan-Taber

Marcus

Stanek

E III

. A randomized controlled trial of prenatal physical activity to prevent gestational diabetes: design and methods. J Womens Health (Larchmt). 2009;18(6):851–859.

27.

Institute of Medicine, Subcommittee on Nutritional Status and Weight Gain During Pregnancy. Nutrition During Pregnancy. Washington, DC: Institute of Medicine; 1990.

28.

Committee on Obstetric Practice. ACOG committee opinion. Exercise during pregnancy and the postpartum period. Number 267, January 2002. American college of obstetricians and gynecologists. Int J Gynaecol Obstet. 2002;77(1):79–81.

29.

Oken

Taveras

Kleinman

Rich-Edwards

Gillman

. Gestational weight gain and child adiposity at age 3 years. Am J Obstet Gynecol. 2007;196(4):322.e1-322.e8.

30.

Stevens-Simon

Roghmann

McAnarney

. Relationship of self-reported prepregnant weight and weight gain during pregnancy to maternal body habitus and age. J Am Diet Assoc. 1992;92(1):85–87.

31.

Nagey

. Validity of self-reported pregravid weight. Ann Epidemiol. 1992;2(5):715–721.

32.

Phelan

Phipps

Abrams

Darroch

Schaffner

Wing

. Randomized trial of a behavioral intervention to prevent excessive gestational weight gain: the fit for delivery study. Am J Clin Nutr. 2011;93(4):772–779.

33.

Institute of Medicine and National Research Council. Weight Gain During Pregnancy: Reexamining the Guidelines. Washington, DC: The National Academies Press; 2009.

34.

Chasan-Taber

Schmidt

Roberts

Hosmer

Markenson

Freedson

. Development and validation of a pregnancy physical activity questionnaire. Med Sci Sports Exerc. 2004;36(10):1750–1760.

35.

Centers for Disease Control and Prevention. Pregnancy nutrition surveillance, nation, summary of trends in maternal health indicators by race/ethnicity, table 20D. http://www.cdc.gov/pednss/pnss_tables/pdf/national_table20.pdf. Updated 2009. Accessed April 25, 2015.

36.

Wolfgang

Viechtbauer

. Conducting Meta-Analyses in R with the Metafor Package. J of Stat Software. 2010;36(3).

37.

R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2014. http://www.R-project.org/. Updated 2014. Accessed September 21, 2014.

38.

Renault

Norgaard

Nilas

. The treatment of obese pregnant women (TOP) study: a randomized controlled trial of the effect of physical activity intervention assessed by pedometer with or without dietary intervention in obese pregnant women. Am J Obstet Gynecol. 2014;210(2):134.e1-134.e9.

39.

Hawkins

Chasan-Taber

Marcus

. Impact of an exercise intervention on physical activity during pregnancy: the behaviors affecting baby and you study. Am J Public Health. 2014;104(10): e74–e81.