Objective:

We conducted this systematic review to support the U.S. Preventive Services Task Force (USPSTF) in updating its 2008 recommendation on counseling to promote and support breastfeeding. Our review addressed three questions: 1) What are the effects of prenatal, peripartum, and postpartum individual- and health care system-level interventions to promote and support breastfeeding on child and maternal health outcomes? 2) What are the effects of interventions on the initiation, duration, and exclusivity of breastfeeding? 3) Are there adverse events associated with interventions to promote and support breastfeeding?

Data Sources:

We performed a search of MEDLINE, PubMed Publisher-Supplied, Cumulative Index for Nursing and Allied Health Literature, Cochrane Central Register of Controlled Trials, and PsycInfo for studies published between January 1, 2008, and September 25, 2015. Studies included in the original USPSTF review were re-evaluated for inclusion. We supplemented searches by examining bibliographies from retrieved articles and consulting outside experts. We searched federal and international trial registries for ongoing trials.

Study Selection:

Two researchers reviewed 2,769 abstracts and 211 articles against the prespecified inclusion criteria. Eligible studies included English-language studies conducted in a developed country that evaluated the effectiveness of an individual- or system-level breastfeeding intervention among pregnant women or mothers of full- or near-term infants. We included randomized or cluster randomized, controlled trials for individual-level interventions and controlled before-after or prospective cohort studies for health system or policy interventions that reported health or breastfeeding outcomes. We conducted dual, independent critical appraisal of all provisionally included studies and abstracted all important study details and results from fair- and good-quality studies. Data were independently abstracted by one reviewer and confirmed by another.

Data Analysis:

We narratively synthesized the results for health outcomes and adverse events. For breastfeeding outcomes, we synthesized the results by population (adults separately from adolescents or young adults) and intervention focus (individual- vs. system-level approaches). Because of the small number of system-level interventions, we report those results narratively and do not pool the data. For individual-level interventions, we conducted random effects meta-analyses using the DerSimonian and Laird method and calculated pooled risk ratios (RRs) for breastfeeding initiation and for any or exclusive breastfeeding at postpartum time points of less than 3 months, 3 to less than 6 months, and 6 months. We explored potential effect modification by various population and intervention characteristics, such as intention to breastfeed and intervention type, and timing through stratified analyses and meta-regression. We generated funnel plots and conducted tests for small-study effects for all pooled analyses.

Results:

We included 52 studies that were reported in 57 publications. Thirty one studies were newly identified while 21 studies were carried forward from the previous review. The included studies were highly variable in terms of the country, study population, intervention and control conditions, specific outcome measures, and timing of measurements.

Infant and maternal health outcomes. Six trials reported inconsistent effects of the interventions on a range of infant health outcomes, such as gastrointestinal illness, otitis media, respiratory tract illness, and health care use. None of the studies reported maternal health outcomes.

Breastfeeding outcomes. On the basis of 43 trials, breastfeeding support and education interventions targeting individuals were associated with a statistically significant higher likelihood of any and exclusive breastfeeding for less than 3 months and at 3 to less than 6 months compared with usual care among adults. Pooled estimates indicated beneficial associations for any breastfeeding for less than 3 months (RR, 1.07 [95% confidence interval (CI), 1.03 to 1.11]; k=26; n=11,588) and at 3 to less than 6 months (RR, 1.11 [95% CI, 1.04 to 1.18]; k=23; n=8,942) and for exclusive breastfeeding for less than 3 months (RR, 1.21 [95% CI, 1.11 to 1.33]; k=22; n=8,246) and at 3 to less than 6 months (RR, 1.20 [95% CI, 1.05 to 1.38]; k=18; n=7,027). At 6 months, individual-level interventions among adults were associated with a 16 percent higher likelihood of exclusive breastfeeding (RR, 1.16 [95% CI, 1.02 to 1.32]; k=17; n=7,690) but not any breastfeeding. Absolute differences in the rates of any breastfeeding ranged from 14.1 percent in favor of the control group to 18.4 percent in favor of the intervention group. The association between individual-level interventions and breastfeeding initiation was not statistically significant based on the pooled point estimate, but the CI did not rule out potential benefit (RR, 1.00 [95% CI, 0.99 to 1.02]; k=14; n=9,428). There was some suggestion that interventions that took place during a combination of prenatal, peripartum, or postpartum time periods were more effective than those that took place only during one time period. There was no indication of effect modification by other intervention characteristics or by population subgroups. All four trials of individual-level interventions among adolescents or young adults reported higher rates of breastfeeding among intervention versus control group participants. There was limited, mixed evidence from well-controlled studies of an association between system-level interventions and rates of breastfeeding.

Adverse events. Two trials among adults reported on adverse events related to a breastfeeding support intervention. One trial found no significant differences between groups in maternal anxiety at 2 weeks. The other reported that a few mothers expressed feelings of anxiety and decreased confidence in their breastfeeding abilities despite breastfeeding going well and therefore discontinued their participation in the peer counseling intervention.

Limitations:

There were a number of threats to internal validity within the included studies. Details regarding the measurement of breastfeeding outcomes, sociodemographic and breastfeeding-related population characteristics, and intervention and usual care characteristics were lacking. Our pooled analyses relied on unadjusted breastfeeding rates and did not control for potential confounding.

Conclusions:

The body of fair- to good-quality evidence related to primary care interventions to support breastfeeding has nearly doubled since the release of the 2009 USPSTF review and recommendation. The updated evidence confirms that breastfeeding support and education provided by professionals and peers to individual women, regardless of the mother's age, is associated with an increase in the duration of any and exclusive breastfeeding. There are limited well-controlled studies examining the effectiveness of system-level policies and practices on rates of breastfeeding, as well as on child health, and none for maternal health.