Background:

Counseling and active behavioral interventions to limit excess gestational weight gain (GWG) during pregnancy may improve health outcomes for women and infants.

Purpose:

To synthesize evidence on the effects of counseling and active behavioral interventions for healthy weight and weight gain during pregnancy for the United States Preventive Services Task Force (USPSTF).

Data Sources:

Cochrane Central Register of Controlled Trials (through March 20, 2020) and Cochrane Database of Systematic Reviews (through March 20, 2020), and MEDLINE (1946 to March 20, 2020), and manually reviewed reference lists, with surveillance through February 5, 2021.

Study Selection:

English-language randomized controlled trials and controlled trials of the effectiveness, benefits, and adverse effects of counseling and active behavioral interventions to limit excess GWG during pregnancy or in women planning pregnancy, including adolescents.

Data Extraction:

One investigator abstracted details about study design, patient population, setting, intervention, followup, and results, reviewed by a second investigator for accuracy. Two investigators independently assessed study quality using methods developed by the USPSTF. Discrepancies were resolved through consensus.

Data Synthesis (Results):

Sixty-eight studies (64 randomized clinical trials and 4 nonrandomized trials) of interventions to limit excess GWG during pregnancy were included. Sixty-seven studies evaluated interventions during pregnancy and one trial evaluated an intervention prior to pregnancy

GWG interventions were associated with small, but statistically significant reductions in risk of gestational diabetes mellitus (GDM) (43 trials; relative risk [RR], 0.87 [95% confidence interval (CI), 0.79 to 0.95]; I²=16.4%; absolute risk difference [ARD], −1.6% [95% CI, −2.5 to −0.7]) and emergency cesarean delivery (14 trials; RR, 0.85 [95% CI, 0.74 to 0.96]; I²=0%; ARD, −2.4% [95% CI, −4.2 to −0.3]). There was no association between GWG interventions versus controls for gestational hypertension (28 trials; RR, 0.87 [95% CI, 0.70 to 1.04]; I²=32.5%; ARD, −0.8% [95% CI, −1.9 to 0.2]), risk of cesarean delivery (34 trials; RR, 0.98 [95% CI, 0.91 to 1.04]; I²=10.8%; ARD, −0.7% [95% CI, −2.4 to 0.8]), preeclampsia (27 trials; RR, 0.98 [95% CI, 0.84 to 1.13], I²=0%; ARD, 0.1% [95% CI, −0.6 to 0.5]), postpartum hemorrhage (9 trials; RR, 1.00 [95% CI, 0.88 to 1.15]; I²=0%; ARD, −0.1% [95% CI, −1.3 to 1.5]), or perineal trauma (8 trials; RR, 0.86 [95% CI, 0.52 to 1.50]; I²=57.0%; ARD, −0.7% [95% CI, −3.2 to 1.7]). Stratified analyses demonstrated statistically significant interactions between effects of GWG interventions on gestational hypertension, high intensity interventions (p=0.006 for interaction) and active interventions (p<0.001 for interaction); and effects on perineal trauma and BMI category (p=0.003 for interaction) and intervention intensity (p=0.003 for interaction); but subgroup effects were not observed for other maternal health outcomes. GWG interventions were not associated with maternal death (2 trials), but data were limited by few trials and low event rates.

Data on interventions to reduce prepregnancy weight were limited. One trial showed a reduction in prepregnancy weight loss associated with a counseling intervention; however, intervention participants gained more weight versus controls during pregnancy (13.2 kg vs. 10.3 kg, p=0.03), with no effect on rates of excess GWG.

For infant health outcomes, GWG interventions were associated with decreased risk of infant macrosomia (25 trials; RR, 0.77 [95% CI, 0.65 to 0.92]; I²=38.3%; ARD, −1.9% [95% CI, −3.3 to −0.7]) and large for gestational age (26 trials; RR, 0.89 [95% CI, 0.80 to 0.99]; I²=0%; ARD, −1.3, [95% CI, −2.3 to −0.3]), but were not associated with risk of preterm birth, respiratory distress syndrome, shoulder dystocia, or neonatal intensive care unit admission. GWG interventions were not associated with neonatal death or stillbirth (11 trials); data were limited by few trials and low event rates. Stratified analyses demonstrated statistically significant interactions between effects of GWG interventions on macrosomia and high intensity interventions (p=0.03 for interaction), but subgroups effects were not observed for other infant health outcomes.

GWG interventions were associated with one kilogram lower weight gain across all prepregnancy weight categories (55 trials; pooled mean difference [MD], −1.02 kg [95% CI, −1.30 to −0.75]; I²=60.3%). High-intensity interventions (12 or more sessions) were associated with greater effects (28 trials; MD, −1.47 kg [95% CI, −1.78 to −1.22]; I²=13.0%) than moderate- (3-11 sessions) (18 trials; MD, −0.32 kg [95% CI, −0.71 to −0.04]; I²=17.6%) or low-intensity interventions (fewer than 2 sessions) (9 trials; MD, −0.64 kg [95% CI, −1.44 to 0.02]; I²=48.4%; p<0.001 for interaction). There was no interaction between intervention type (active vs. counseling only) or baseline BMI category and effects on GWG. Interventions were also associated with a lower likelihood of gaining gestational weight in excess of the Institute of Medicine (IOM) recommendations (39 trials; RR, 0.83 [95% CI, 0.77 to 0.89]; I²=63.8%; ARD, −7.6% [95% CI, −11.0 to −4.6]), with greater effect for active (p=0.003 for interaction) and high intensity interventions (p<0.001 for interaction); there was no interaction between BMI category and effects on likelihood of excess weight gain. GWG interventions were not associated with increased likelihood of adherence to IOM recommendations for GWG (i.e., neither gaining excessive weight nor failing to gain sufficient weight) or postpartum weight retention at less than 6 or 6 months, but were associated with reduced postpartum weight retention at 12 months (10 trials; MD, −0.63 kg [95% CI, −1.44 to −0.01]; I²=65.5%).

Data on harms were limited. Twelve studies of the effects of GWG interventions on maternal anxiety and depression showed mixed results. The pooled estimate of the effect of maternal GWG interventions on infants small for gestational age was not statistically significant (20 trials; RR, 0.94 [95% CI, 0.80 to 1.10]; I²=0%; ARD, −0.4% [95% CI, −1.7 to 1.0]). Stratified analyses demonstrated a statistically significant interaction between SGA and intervention intensity (p=0.04 for interaction), but not BMI category or intervention type.

Limitations:

Restricted to English-language articles, statistical heterogeneity in pooled analyses, limited evidence on infant health outcomes and harms of interventions, and most trials had some methodological limitations. Data were lacking on effectiveness of prepregnancy interventions, and on GWG interventions in pregnant adolescents and women with advanced maternal age.

Conclusions:

Counseling and active behavioral interventions to limit GWG in pregnant women are associated with a modestly decreased risk of GDM, emergency cesarean delivery, macrosomia, and large for gestational age. GWG interventions are also associated with modest reductions in weight gain and decreased likelihood of exceeding IOM recommendations for GWG. Effects of these interventions on mean GWG are slightly more pronounced for high intensity interventions.