Abstract
Introduction:
Social support is a modifiable social determinant of health that shapes breastfeeding outcomes and may contribute to racial and ethnic breastfeeding disparities. This study characterizes the relationship between social support and early breastfeeding.
Methods:
This is a cross-sectional analysis of baseline data collected in 2019‒2021 for an RCT. Social support was measured using the ENRICHD Social Support Instrument (ESSI). Outcomes, collected via self-report, included (1) early breastfeeding within the first 21 days of life (2) planned breastfeeding duration, and (3) confidence in meeting breastfeeding goal. Each outcome was modeled using proportional odds regression, adjusting for covariates. Analysis was conducted in 2021‒2022.
Results:
Self-reported race and ethnicity among 883 mothers was: 50% Hispanic, 17% Black, 23% White, and 10% other. A large proportion (88%) of mothers were breastfeeding. Most breastfeeding mothers (82%) planned to breastfeed for at least 6 months, with over half (58%) planning to continue for 12 months or more. Most women (65%) were “confident” or “very confident” in meeting their breastfeeding duration goal. In adjusted models, perceived social support was associated with planned breastfeeding duration (p=0.042), but not with early breastfeeding (p=0.873) or confidence in meeting breastfeeding goal (p=0.427). Among the covariates, maternal depressive symptoms were associated with lower breastfeeding confidence (p<0.001).
Conclusions:
The associations between perceived social support and breastfeeding outcomes are nuanced. In this sample of racially and ethnically diverse mothers, social support was associated with longer planned breastfeeding duration but not with early breastfeeding or breastfeeding confidence.
INTRODUCTION
Breastfeeding offers many health benefits to mothers and children as well as economic and environmental advantages for society as a whole, yet these benefits are not evenly experienced by all racial and ethnic groups.1‒6 The American Academy of Pediatrics and WHO recommend exclusive breastfeeding for the first 6 months of life followed by appropriate solid foods alongside continued breastfeeding up to 2 years of life or beyond, as mutually desired by mother and child.7,8 Although breastfeeding initiation in the U.S. reached 83.2% in 2019, only 24.9% of infants were exclusively breastfed through 6 months with disparities evident: 19.1% of non-Hispanic Black infants were exclusively breastfeeding at 6 months compared to 23.5% of Hispanic infants and 26.9% of non-Hispanic White infants.9 To achieve the Healthy People 2030 goal of increasing the proportion of infants who are exclusively breastfed through 6 months of age, a greater understanding of the factors that support breastfeeding among a wide range of communities is needed.10
Social determinants of health—the conditions in which people are born, grow, work, learn, play, and live—are potentially modifiable factors that have been previously associated with breastfeeding outcomes.3,11,12 Social support is one such modifiable social determinant of health that could play an important role in shaping breastfeeding outcomes. The American Psychological Association defines social support as assistance or comfort to help cope with biological, psychological, and social stressors.13 Although intuition suggests a positive association between social support and breastfeeding outcomes, several recent studies have found contradictory results.14,15 One potential explanation is that these studies have not included measures of social support that extend beyond partners and family members to include support from friends and neighbors.16,17 Because social support is a modifiable driver of breastfeeding decisions, it is especially important to continue exploring the relationship between social support and breastfeeding among racial and ethnic groups that have lower rates of breastfeeding.18‒23 Understanding how a broad conceptualization of social support is associated with breastfeeding outcomes will inform whether interventions that enhance social support for new mothers could improve breastfeeding rates and reduce breastfeeding disparities.
The purpose of this study was to characterize the relationship between a broad conception of social support and early breastfeeding, planned breastfeeding duration, and confidence in meeting breastfeeding goal among a racially and ethnically diverse population. The secondary objective was to determine whether the associations between social support and these breastfeeding outcomes varied by maternal race and ethnicity.
METHODS
This is a cross-sectional analysis using baseline data from the Greenlight Plus Study: Approaches to Early Obesity Prevention. The Greenlight Plus Study is a multi-center RCT, described in Appendix 1.24 All participants signed an informed consent document prior to study participation. The IRB at Vanderbilt University Medical Center served as the single IRB for the Greenlight Plus study, coordinating approvals from the IRB at each of the other sites. The Greenlight Plus Study is registered with the national Clinical Trials Registry (NCT04042467 at clinicaltrials.gov).
Study Sample
To be eligible for the Greenlight Plus Study, children had to be born after 34 weeks gestation, have a weight >3rd percentile based on WHO growth curves, and not have any chronic medical problem that could affect weight gain. Full inclusion criteria are listed in Appendix 1. The Greenlight Plus Study enrolled 900 caregiver-infant dyads who were recruited from 6 networks of newborn nurseries/primary care clinics (Duke University, University of Miami, New York University, University of North Carolina, Stanford University, and Vanderbilt University Medical Center). At all 6 sites, most children are insured by Medicaid or other public insurance, and a large proportion of caregivers self-identify as non-Hispanic Black or Hispanic. Most participants were recruited through a phone call following their first newborn clinic visit. Baseline surveys were collected over the phone at the time of the recruitment phone call or during a subsequent phone call, depending on caregiver availability. All baseline survey data were collected within the first 21 days of life. Research assistants extracted some data directly from the electronic health record (e.g., birth weight, date of birth). Study enrollment occurred from October 2019 through August 2021.
Measures
Social support was measured using the first 6 items on the ENRICHD Social Support Instrument (ESSI), which is a previously validated measure.25,26 The ESSI is a 7-item self-report measure that incorporates 4 domains of social support: emotional (caring), instrumental (tangible goods and services), informational (information provided in times of stress), and appraisal (communication of information relevant to self-evaluation). The seventh item (“Are you currently married or living with a partner?”) was included in this study as a separate covariate, based on previously published approaches to scoring the ESSI. The ESSI score is reported as a sum of 6 items on a continuous scale for a total ranging from 6 to 30. This continuous score was the primary predictor of interest in each of the statistical models, described below. A higher ESSI score indicates a higher level of social support. To compare with other studies in the literature, the ESSI is also described based on a previously identified cut point: a total score ≤18 and 2 or more individual items with a score ≤3 as “low social support”.25,27‒29
Outcomes included (1) early breastfeeding, (2) planned breastfeeding duration, and (3) confidence in meeting breastfeeding goal. Each outcome was collected through self-report using questions modeled after the Infant Feeding Practices Study II Neonatal Questionnaire.30,31 For early breastfeeding, mothers were asked at enrollment “What type of food does [child first name] eat?”. Responses were categorized as exclusive breastfeeding, breast and formula feeding (mixed feeding), or no breastfeeding (exclusive formula feeding). Only mothers who indicated breastfeeding at baseline (exclusive breastfeeding or mixed feeding) were asked about planned breastfeeding duration. These mothers were asked “How old do you think your baby will be when you completely stop breastfeeding?”. Mothers responded with an integer number of months, selected “I don’t know”, or selected “As long as possible (but can’t say a number)”. The number of months was included as a continuous outcome in the primary models. For descriptive purposes, numerical responses were also categorized into the following categories: <6 months, 6‒11 months, 12‒23 months, or ≥24 months. Only mothers who indicated breastfeeding at baseline and indicated a breastfeeding duration goal were asked about their confidence in meeting their breastfeeding goal. These mothers were asked “How confident are you that you will be able to breastfeed until the baby is the age you marked in the previous question?”. Responses were recorded on a Likert-type scale: “1 - not at all confident”, “2 - slightly confident”, “3 - somewhat confident”, “4 - confident”, and “5 - very confident”.
Covariates were selected a priori based on their theoretical potential to be confounders of the associations of interest. Each covariate was assessed through self-report. Covariates included: household income, household WIC status, number of children in the household, maternal race and ethnicity, maternal education, maternal employment, maternal marital status, maternal depressive symptoms (measured with the PHQ-9),32 maternal age, maternal pre-pregnancy BMI, delivery method, infant age, and study site.33 Additional details on covariates can be found in Appendix 2.
Statistical Analysis
Patient characteristics were stratified by baseline feeding type (exclusive breastfeeding, mixed feeding, and exclusive formula feeding), using the median with IQR for continuous variables and count with percentage for categorical variables.
Three separate proportional odds logistic regression models were fit to estimate the association between social support and the ordinal outcomes: (1) baseline feeding type (exclusive formula feeding, mixed feeding, and exclusive breastfeeding), (2) planned breastfeeding duration (in months) and (3) confidence in meeting the breastfeeding duration goal (5-point Likert-type scale from not at all confident to very confident). To situate the results in the broader context of social determinants of health, how the sociodemographic factors measured as covariates affected the breastfeeding outcomes was also observed. All regression models were adjusted for the covariates listed above. Quantitative covariates (BMI, maternal age, child age, PHQ-9 score, and ESSI score) were included in the regression models using flexible restricted cubic spline to allow for potentially non-linear relationships with the cumulative log odds outcomes.34,35 Because proportional odds logistic regression models were used, estimates of association are reported on the OR scale (i.e., the OR of higher outcome values associated with changes in social support and PHQ-9 score).
The extent to which associations of social support with the 3 outcomes were modified by maternal race and ethnicity was also examined by adding a social support by maternal race and ethnicity interaction term to each model. Missing data were multiply imputed with chained equations using a predictive mean matching strategy.36 Twenty-five imputation datasets were generated and analyzed, and estimates from the imputation-specific analyses were combined with Rubin’s Rule.37 All analyses were conducted with R software version 3.6.3, and specifically the rms library.38,39
RESULTS
For the Greenlight Plus RCT, 3,224 caregivers were approached and 900 caregiver-infant dyads were enrolled. Of those dyads, 17 were excluded from the present study about breastfeeding because the caregiver was not the biological mother, resulting in 883 mother-infant dyads included in analyses.
Characteristics of mother-infant dyads in the Greenlight Plus Study are shown in Table 1. About half (49%) of the mothers lived in households with an income <$50,000/year, and 64% of the mothers received WIC benefits for themselves, their infant, or both. Median maternal age was 30.1 (IQR=25.6‒34.2) years, and median maternal BMI was 26.0 (IQR=22.3‒31.4) kg/m2. Median infant age at the time of survey administration was 10 (IQR=7‒14) days. Median ESSI score was 29 (IQR=26‒30), with 53 mothers (6%) falling into the low social support category. Cronbach’s alpha for the ESSI in the current dataset was 0.85 (95% CI=0.83, 0.86).
Table 1.
Baseline Characteristics of Mother‒Infant Dyads, Overall and by Baseline Feeding Type
| Characteristics | Total (N=883) | Formula only (N=102) | Breast milk and formula (N=410) | Breast milk only (N=371) |
|---|---|---|---|---|
| Household income | ||||
| <$20,000 | 216 (24%) | 43 (42%) | 109 (27%) | 64 (17%) |
| $20,000 to $49,999 | 219 (25%) | 33 (32%) | 106 (26%) | 80 (22%) |
| $50,000 to $99,999 | 97 (11%) | 4 (4%) | 29 (7%) | 64 (17%) |
| ≥$100,000 | 118 (13%) | 3 (3%) | 24 (6%) | 91 (25%) |
| I don’t know | 233 (26%) | 19 (19%) | 142 (35%) | 72 (19%) |
| Household WIC status | ||||
| No | 319 (36%) | 19 (19%) | 94 (23%) | 206 (56%) |
| Yes | 564 (64%) | 83 (81%) | 316 (77%) | 165 (44%) |
| Number of children in household | ||||
| 1 | 333 (38%) | 33 (32%) | 147 (36%) | 153 (41%) |
| 2 | 263 (30%) | 27 (26%) | 112 (27%) | 124 (33%) |
| ≥3 | 287 (33%) | 42 (41%) | 151 (37%) | 94 (25%) |
| Maternal race/ethnicity | ||||
| Black, non-Hispanic | 151 (17%) | 41 (40%) | 56 (14%) | 54 (15%) |
| Hispanic | 443 (50%) | 41 (40%) | 266 (65%) | 136 (37%) |
| Other, non-Hispanic | 89 (10%) | 4 (4%) | 36 (9%) | 49 (13%) |
| White, non-Hispanic | 200 (23%) | 16 (16%) | 52 (13%) | 132 (36%) |
| Maternal education | ||||
| <High school graduate | 195 (22%) | 28 (27%) | 126 (31%) | 41 (11%) |
| High school graduate, <college degree | 382 (43%) | 63 (62%) | 185 (45%) | 134 (36%) |
| College degree or higher | 306 (35%) | 11 (11%) | 99 (24%) | 196 (53%) |
| Maternal employment | ||||
| Full time employed | 341 (39%) | 38 (37%) | 119 (29%) | 184 (50%) |
| Part time employed | 70 (8%) | 4 (4%) | 45 (11%) | 21 (6%) |
| Unemployed | 471 (53%) | 60 (59%) | 245 (60%) | 166 (45%) |
| Maternal marital status | ||||
| Married | 383 (43%) | 19 (19%) | 150 (37%) | 214 (58%) |
| Member of unmarried couple living together | 231 (26%) | 30 (30%) | 118 (29%) | 83 (22%) |
| Divorced, separated, or widowed | 22 (2%) | 7 (7%) | 9 (2%) | 6 (2%) |
| Single, never married | 246 (28%) | 45 (45%) | 133 (32%) | 68 (18%) |
| Maternal depressive symptoms (PHQ-9 score) | 2 [0, 4] | 1 [0, 4] | 2 [0, 4] | 2 [1, 4] |
| Maternal age (years) | 30.0 [25.6, 34.2] | 27.4 [23.4, 31.6] | 30.1 [25.5, 34.7] | 30.7 [26.7, 34.2] |
| Infant age (days) | 10 [7, 14] | 10 [7, 13] | 10 [7, 14] | 10 [6, 14] |
| Maternal BMI (kg/m2) | 26.0 [22.3, 31.4] | 26.5 [21.8, 35.0] | 27.1 [23.0, 32.6] | 24.9 [22.0, 30.2] |
| Number of children in household | 2 [1, 3] | 2 [1, 3] | 2 [1, 3] | 2 [1, 3] |
| Delivery method | ||||
| Caesarean section | 242 (27%) | 34 (33%) | 112 (27%) | 96 (26%) |
| Vaginal delivery | 641 (73%) | 68 (67%) | 298 (73%) | 275 (74%) |
| EESI categories | ||||
| Low social support (<18 total and <2 on 3+ questions) | 53 (6%) | 5 (5%) | 37 (9%) | 11 (3%) |
| Adequate social support | 825 (94%) | 97 (95%) | 370 (91%) | 358 (97%) |
| EESI score | 29 [26, 30] | 29 [25, 30] | 29 [24, 30] | 30 [27, 30] |
Notes: Categorical variables presented as count with percentages and continuous variables with median and interquartile range.
ESSI, ENRICHD Social Support Instrument; PHQ-9, Patient Health Questionnaire-9; WIC, Supplemental Nutrition Program for Women, Infants, and Children.
Breastfeeding outcomes overall and by race and ethnicity are shown in Table 2. A large proportion (88%) of mothers were doing some breastfeeding at the time of survey collection. Most mothers who were breastfeeding (82%) planned to breastfeed for at least 6 months, with over half (57%) planning to continue for 12 months or more. Most women (69%) were “confident” or “very confident” that they could meet their breastfeeding duration goal.
Table 2.
Breastfeeding Outcomes, Overall and by Race/Ethnicity
| Variables | Total | Black, non-Hispanic | Hispanic | Other, non-Hispanic | White, non-Hispanic |
|---|---|---|---|---|---|
| Early breastfeeding (N) | 883 | 151 | 443 | 89 | 200 |
| Formula only | 102 (12%) | 41 (27%) | 41 (9%) | 4 (4%) | 16 (8%) |
| Breast milk and formula | 410 (46%) | 56 (37%) | 266 (60%) | 36 (40%) | 52 (26%) |
| Breast milk only | 371 (42%) | 54 (36%) | 136 (31%) | 49 (55%) | 132 (66%) |
| Planned breastfeeding durationa (N) | 781 | 110 | 402 | 85 | 184 |
| <6 months | 50 (6%) | 11 (10%) | 24 (6%) | 4 (5%) | 11 (6%) |
| 6‒11 months | 191 (24%) | 37 (34%) | 86 (21%) | 15 (18%) | 53 (29%) |
| 12‒23 months | 379 (49%) | 44 (40%) | 197 (49%) | 40 (47%) | 98 (53%) |
| ≥24 months | 70 (9%) | 10 (9%) | 36 (9%) | 13 (15%) | 11 (6%) |
| I don’t know | 56 (7%) | 5 (5%) | 34 (8%) | 9 (11%) | 8 (4%) |
| As long as possible (but can’t say a number) | 35 (4%) | 3 (3%) | 25 (6%) | 4 (5%) | 3 (2%) |
| Confidence in meeting breastfeeding goalb (N) | 725 | 105 | 368 | 76 | 176 |
| Not at all confident | 29 (4%) | 2 (2%) | 24 (7%) | 1 (1%) | 2 (1%) |
| Slightly confident | 51 (7%) | 2 (2%) | 41 (11%) | 1 (1%) | 7 (4%) |
| Somewhat confident | 137 (19%) | 19 (18%) | 55 (15%) | 19 (25%) | 44 (25%) |
| Confident | 237 (33%) | 36 (35%) | 119 (33%) | 26 (35%) | 56 (32%) |
| Very confident | 262 (37%) | 44 (43%) | 124 (34%) | 28 (37%) | 66 (38%) |
Notes: Categorical variables presented as count with percentages.
Patients who were not breastfeeding were excluded.
Patients who were not breastfeeding or responded as “I don’t know” when asked to give a planned breastfeeding duration were excluded.
In 3 separate covariate adjusted regression analyses, perceived social support was not associated with early breastfeeding (p=0.873) or confidence in meeting breastfeeding goal (p=0.940). However, perceived social support was associated with planned breastfeeding duration (p=0.042). Higher levels of perceived social support increased the odds of planning to breastfeed for longer up to an ESSI score of about 25, at which point the relationship appeared to level off (Figure 1). As an example, a mother with an ESSI score of 10 had about half the odds of planning to breastfeed for longer compared to mothers with an ESSI score of 18 (the threshold for adequate social support) (OR=0.57, 95% CI=0.36, 0.92). Further, a mother with an ESSI score of 28 had over 1.66 times the odds of planning to breastfeed for longer compared to mothers with an ESSI score of 18 (OR=1.69, 95% CI=1.12, 2.47). Appendix 3 provides the data table showing ORs for the full range of ESSI scores. When evaluating for the presence of effect modification by race and ethnicity, the p-value of the interaction term between social support and the race/ethnicity variable was not significant (>0.05) in each of the 3 models: early breastfeeding (p=0.171), planned breastfeeding duration (p=0.773), and confidence in meeting breastfeeding goal (p=0.517).
Figure 1.
Association between perceived social support and planned breastfeeding duration.
Notes: This figure plots ORs from the fully adjusted proportional odds regression model against maternal perceived social support, measured by the ENRICHD Social Support Instrument (ESSI). Higher scores on the ESSI indicate higher levels of perceived social support. AORs are shown compared to a reference value of ESSI=18, the published cut point for “low social support.” AORs are shown with associated 95% CIs. The overall p-value of 0.04 is based on a likelihood ratio test, as social support is modeled non-linearly.
ENRICHD, Enhancing Recovery in Coronary Heart Disease.
Depressive symptoms were associated with mothers’ confidence in meeting their breastfeeding goal (p<0.001). Mothers with more depressive symptoms (higher PHQ-9 score) had lower confidence in meeting their breastfeeding goal (Figure 2). Appendix 4 provides a data table showing ORs for the full range of PHQ-9 scores. Depressive symptoms were not associated with early breastfeeding (p=0.534) or planned breastfeeding duration (p=0.595).
Figure 2.
Association between maternal depressive symptoms and confidence in meeting breastfeeding goal.
Notes: This figure plots ORs from the fully adjusted proportional odds regression model against maternal depressive symptoms, measured by the Patient Health Questionnaire-9 (PHQ-9). Higher scores on the PHQ-9 indicate higher levels of depressive symptoms. AORs are shown compared to a reference value of PHQ-9=0. AORs are shown with associated 95% CIs. The overall p-value below 0.001 is based on a likelihood ratio test, as depressive symptoms are modeled non-linearly.
Full model output showing all covariate associations with early breastfeeding, planned breastfeeding duration, and confidence in meeting breastfeeding goal is shown in Appendix 5, 6, and 7, respectively.
DISCUSSION
Despite an intuitive sense that social support should broadly improve breastfeeding outcomes, the associations between social support and the three breastfeeding outcomes of interest were variable. While higher levels of social support were associated with longer planned breastfeeding duration, this study was unable to detect an association between social support and early breastfeeding or between social support and breastfeeding confidence. This variability may be a significant factor when considering interventions to promote breastfeeding. Tools to enhance social support may or may not be useful depending on the specific target outcome of an intervention. For example, to improve early breastfeeding and breastfeeding confidence in particular, interventions focused on other influencers of breastfeeding may be more helpful than interventions focused on social support. Additionally, there was a threshold effect for the association between social support and longer planned breastfeeding duration, with an apparent strong association only up to an ESSI score of 25. This suggests that interventions to enhance social support may be particularly relevant among mothers with low social support. However, once a mother reaches a certain threshold level of social support, further intervention to enhance social support may produce diminishing returns in lengthening breastfeeding duration.
The finding that social support is associated with longer planned breastfeeding duration is important given evidence that intended breastfeeding duration is strongly associated with actual breastfeeding duration.40 Importantly, a previous study involving a low income and racially and ethnically diverse population found that although longer intended breastfeeding duration predicted longer actual breastfeeding duration, mothers were unlikely to reach their breastfeeding duration goals.41 Thus, in tandem with interventions that lengthen planned breastfeeding duration, interventions to help mothers reach their goals are needed. As data from later time points in the Greenlight Plus Study become available, examining whether higher levels of social support are associated with greater likelihood of achieving breastfeeding duration goals could reveal whether interventions targeting social support for mothers could both lengthen planned breastfeeding duration and help mothers reach their longer breastfeeding duration goals.
Compared to previous evaluations of social support and breastfeeding outcomes, the current analysis accounted for a wider range potential confounders (maternal BMI, mode of delivery, parity), was strengthened by a more diverse sample of mothers, and included a general measure of social support that incorporated a wider range of social support domains.14,42 Developing an instrument that robustly measures each domain of social support specifically in the context of breastfeeding would reveal more targeted information for developing interventions to improve breastfeeding outcomes. This future research should include analyses that consider health system factors (e.g., lactation support services, hospital policies) and structural factors (e.g., workplace policies, breastmilk substitute marketing), which may influence the relationship between social support and breastfeeding outcomes.6,43 With the well-described disparities in breastfeeding by race and ethnicity, future research should also more directly evaluate the potential contribution of systematic racism to social support and breastfeeding outcomes.2,44
Although this study did not set out to investigate the relationship between depressive symptoms and breastfeeding outcomes, higher levels of maternal depressive symptoms were a significant factor associated with lower breastfeeding confidence. Despite extensive research on the topic, the exact nature of the depression-breastfeeding relationship remains unclear, with many studies coming to conflicting conclusions about the existence and direction of the relationship between maternal depression and various breastfeeding outcomes.45,46 This study found that when using a flexible, non-linear association between depressive symptoms and confidence in meeting breastfeeding goal, the association appeared stronger for lower (<5) PHQ-9 scores. This suggests that even a small number of minor depressive symptoms could decrease breastfeeding confidence. Because mothers with few, minor depressive symptoms would not be flagged as needing additional support through postpartum depression screening, it is important that clinicians recognize even minor depressive symptoms as a risk factor for low breastfeeding confidence and early breastfeeding cessation.
Limitations
There are several limitations to this study. First, all mothers in the sample attended pediatric clinics associated with academic medical centers. Therefore, these results may not be representative of other settings. The sample was limited to English- and Spanish-speaking mothers, preventing generalizability to non-English- and/or Spanish-speaking mothers. Mothers who did not attend their first clinic visit within 21 days of life were also excluded from the study. It is possible that mothers unable to attend this early life appointment experience a lower level of social support. This sample had a high rate of early breastfeeding (88%). While this is comparable to national averages (84% every year between 2015‒2018), the high rate of early breastfeeding reported by mothers in this study may have led to inadequate variation in the outcome to detect associations with social support. Measurement was limited by a lack of information about the proportion of breast milk and formula used by mothers in the mixed feeding group. Heterogeneity within this group, with some mothers using much more breast milk than others, could limit differences seen between groups. Additionally, the question “How old do you think your baby will be when you completely stop breastfeeding?” does not directly ask about a breastfeeding “goal,” and responses may have been influenced by barriers to breastfeeding that the mother has or expects to encounter. Therefore, mothers’ desired breastfeeding duration, or goal, may be longer than responses to this question indicate. Measurement of the outcomes was also limited by concerns related to participant burden in the main RCT. As such, a multi-item measure of breastfeeding confidence and a breastfeeding-specific social support scale were not included. Future research should consider more in-depth measures of these important constructs. Because data were drawn from surveys, self-report bias and social desirability bias are possible, especially given stigma surrounding both breastfeeding and formula feeding. Finally, as with all cross-sectional, observational studies, causation cannot be inferred and the potential for residual confounding should be considered.
CONCLUSIONS
In a large sample of racially and ethnically diverse mothers, social support was associated with longer planned breastfeeding duration, but was not associated with early breastfeeding or breastfeeding confidence. Given the extensive health benefits of breastfeeding for both mothers and children, identifying modifiable factors, such as social support, as targets for lengthening breastfeeding duration is essential for improving public health. Including racially and ethnically diverse mothers in future breastfeeding research and interventions is necessary to reduce preventable disparities in breastfeeding outcomes and eliminate health inequities that may be related to breastfeeding disparities.
Supplementary Material
ACKNOWLEDGMENTS
The authors would like to extend their sincere gratitude to the research coordinators and staff, the community stakeholders, and especially the patients and caregivers who participated in this study for their time and contributions. This work was supported by the Patient Centered Outcomes Research Institute (PCORI) [contract number AD-2018C1–11238]. Study data were collected and managed using REDCap electronic data capture tools hosted at Vanderbilt University Medical Center, and supported by NCATS/NIH, grant number: UL1 TR000445. The funders of this study had no role in the study design; the collection, analysis, and interpretation of data; the writing of the report; or the decision to submit the report for publication. The IRB at Vanderbilt University Medical Center served as the single IRB for the Greenlight Plus study, coordinating approvals from the IRB at each of the other sites (IRB number 190311).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Preliminary data from this paper were presented at the 2022 Pediatric Academic Societies Meeting.
No financial disclosures were reported by the authors of this paper.
Dr. Flower previously reviewed abstracts for PCORI-funded studies unrelated to this study. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
REFERENCES
- 1.Centers for Disease Control and Prevention. Rates of Any and Exclusive Breastfeeding by Sociodemographics among Children Born in 2018 https://www.cdc.gov/breastfeeding/data/nis_data/rates-any-exclusive-bf-socio-dem-2018.html. Accessed December 2, 2021.
- 2.Louis-Jacques A, Deubel TF, Taylor M, Stuebe AM. Racial and ethnic disparities in U.S. breastfeeding and implications for maternal and child health outcomes. Semin Perinatol 2017;41(5):299–307. 10.1053/j.semperi.2017.04.007. [DOI] [PubMed] [Google Scholar]
- 3.Odar Stough C, Khalsa AS, Nabors LA, Merianos AL, Peugh J. Predictors of Exclusive Breastfeeding for 6 Months in a National Sample of US Children. Am J Health Promot 2019;33(1):48–56. 10.1177/0890117118774208. [DOI] [PubMed] [Google Scholar]
- 4.Bartick M, Reinhold A. The burden of suboptimal breastfeeding in the United States: a pediatric cost analysis. Pediatrics 2010;125(5):e1048–e1056. 10.1542/peds.2009-1616. [DOI] [PubMed] [Google Scholar]
- 5.Bartick MC, Schwarz EB, Green BD, Jegier BJ, Reinhold AG, Colaizy TT, et al. Suboptimal breastfeeding in the United States: Maternal and pediatric health outcomes and costs. Matern Child Nutr 2017;13(1). 10.1111/mcn.12366. [DOI] [PMC free article] [PubMed]
- 6.Rollins NC, Bhandari N, Hajeebhoy N, et al. Why invest, and what it will take to improve breastfeeding practices? Lancet 2016;387(10017):491–504. 10.1016/S0140-6736(15)01044-2. [DOI] [PubMed] [Google Scholar]
- 7.WHO. Breastfeeding https://www.who.int/health-topics/breastfeeding#tab=tab_2. Published 2021. Accessed December 2, 2021.
- 8.Meek JY, Noble L, Breastfeeding So. Policy Statement: Breastfeeding and the Use of Human Milk. Pediatrics 2022;150(1). 10.1542/peds.2022-057988. [DOI] [PubMed] [Google Scholar]
- 9.Centers for Disease Control and Prevention. Breastfeeding Among U.S. Children Born 2011–2018, CDC National Immunization Survey https://www.cdc.gov/breastfeeding/data/nis_data/results.html. Published 2021. Accessed December 2, 2021.
- 10.Office of Disease Prevention and Health Promotion OotASfH, Office of the Secretary, and U.S. Department o health and Human Services,. Healthy People 2030: Increase the proportion of infants who are breastfed exclusively through age 6 months - MICH-15
- 11.Cohen SS, Alexander DD, Krebs NF, et al. Factors Associated with Breastfeeding Initiation and Continuation: A Meta-Analysis. J Pediatr 2018;203:190–196.e21. 10.1016/j.jpeds.2018.08.008. [DOI] [PubMed] [Google Scholar]
- 12.Centers for Disease Control and Prevention. About Social Determinants of Health (SDOH) 2021.
- 13.American Psychological Association. American Psychological Association Dictionary of Psychology https://dictionary.apa.org/social-support. Published 2022. Accessed May 18, 2022.
- 14.Kummer L, Duke N, Davis L, Borowsky I. Association of Social and Community Factors with U.S. Breastfeeding Outcomes. Breastfeed Med 2020;15(10):646–654. 10.1089/bfm.2020.0083. [DOI] [PubMed] [Google Scholar]
- 15.White MJ, Kay MC, Truong T, Green CL, Yin HS, Flower KB, et al. Racial and Ethnic Differences in Maternal Social Support and Relationship to Mother-Infant Health Behaviors. Acad Pediatr 2022. 10.1016/j.acap.2022.02.008. [DOI] [PMC free article] [PubMed]
- 16.American Psychological Association. APA Dictionary of Psychology: social support 2020.
- 17.Health Behavior and Health Education: Theory, Research, and Practice In: Rimer Glanz K, Barbara K, Viswanath K, editor. 4th edition. [Google Scholar]
- 18.DeVane-Johnson S, Giscombe CW, Williams R 2, Fogel C, Thoyre S. A Qualitative Study of Social, Cultural, and Historical Influences on African American Women’s Infant-Feeding Practices. J Perinat Educ 2018;27(2):71–85. 10.1891/1058-1243.27.2.71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hohl S, Thompson B, Escareno M, Duggan C. Cultural Norms in Conflict: Breastfeeding Among Hispanic Immigrants in Rural Washington State. Matern Child Health J 2016;20(7):1549–1557. 10.1007/s10995-016-1954-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Freeman A Skimmed: breastfeeding, race, and injustice Stanford University Press. [Google Scholar]
- 21.Jones KM, Power ML, Queenan JT, Schulkin J. Racial and ethnic disparities in breastfeeding. Breastfeed Med 2015;10(4):186–196. 10.1089/bfm.2014.0152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Oniwon O, Tender JA, He J, Voorhees E, Moon RY. Reasons for Infant Feeding Decisions in Low-Income Families in Washington, DC. J Hum Lact 2016;32(4):704–710. 10.1177/0890334416653739. [DOI] [PubMed] [Google Scholar]
- 23.Furman L, Killpack S, Matthews L, Davis V, O’Riordan MA. Engaging Inner-City Fathers in Breastfeeding Support. Breastfeed Med 2016;11(1):15–20. 10.1089/bfm.2015.0092. [DOI] [PubMed] [Google Scholar]
- 24.Sanders LM, Perrin EM, Yin HS, Bronaugh A, Rothman RL, Greenlight Study T. “Greenlight study”: a controlled trial of low-literacy, early childhood obesity prevention. Pediatrics 2014;133(6):e1724–e1737. 10.1542/peds.2013-3867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Enhancing recovery in coronary heart disease patients (ENRICHD): study design and methods. The ENRICHD investigators. Am Heart J 2000;139(1 Pt 1):1–9. 10.1016/S0002-8703(00)90301-6. [DOI] [PubMed] [Google Scholar]
- 26.Gottlieb BH, Bergen AE. Social support concepts and measures. Journal of Psychosomatic Research 2010;69(5):511–520. 10.1016/j.jpsychores.2009.10.001. [DOI] [PubMed] [Google Scholar]
- 27.Bucholz EM, Strait KM, Dreyer RP, et al. Effect of low perceived social support on health outcomes in young patients with acute myocardial infarction: results from the VIRGO (Variation in Recovery: Role of Gender on Outcomes of Young AMI Patients) study. J Am Heart Assoc 2014;3(5):e001252. 10.1161/JAHA.114.001252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Vaglio J Jr Conard M, Poston WS, et al. Testing the performance of the ENRICHD Social Support Instrument in cardiac patients. Health Qual Life Outcomes 2004;2:24. 10.1186/1477-7525-2-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Investigators E Enhancing recovery in coronary heart disease (ENRICHD): baseline characteristics. Am J Cardiol 2001;88(3):316–322. 10.1016/S0002-9149(01)01652-6. [DOI] [PubMed] [Google Scholar]
- 30.Fein SB, Labiner-Wolfe J, Shealy KR, Li R, Chen J, Grummer-Strawn LM. Infant Feeding Practices Study II: study methods. Pediatrics 2008;122(Suppl 2):S28–S35. 10.1542/peds.2008-1315c. [DOI] [PubMed] [Google Scholar]
- 31.Centers for Disease Control and Prevention. Questionnaires: Breastfeeding and Infant Feeding Practices 2021.
- 32.Wang L, Kroenke K, Stump TE, Monahan PO. Screening for perinatal depression with the Patient Health Questionnaire depression scale (PHQ-9): A systematic review and meta-analysis. General Hospital Psychiatry 2021;68:74–82. 10.1016/j.genhosppsych.2020.12.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001;16(9):606–613. 10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Durrleman S, Simon R. Flexible regression models with cubic splines. Statistics in Medicine 1989;8(5):551–561. 10.1002/sim.4780080504. [DOI] [PubMed] [Google Scholar]
- 35.Harrell FE. Regression modeling strategies : with applications to linear models, logistic regression, and survival analysis New York: Springer; 2001. 10.1007/978-1-4757-3462-1. [DOI] [Google Scholar]
- 36.Raghunathan TW LJ, Van Hoewyk J, Solenbeger P. A multivariate technique for multiply imputing missing values using a sequence of regression models. Survey Methodology 2001;27:85–95. [Google Scholar]
- 37.Rubin DB. Multiple Imputation after 18+ Years. Journal of the American Statistical Association 1996;91(434):473–489. 10.1080/01621459.1996.10476908. [DOI] [Google Scholar]
- 38.R Core Team. R: A language and environment for statistical computing. In. Vienna, Austria: R Foundation for Statistical Computing; 2021.
- 39.Harrell FE. rms: Regression Modeling Strategies. R package version 6.1–0 In; 2020.
- 40.Nelson JM, Li R, Perrine CG, Scanlon KS. Changes in mothers’ intended duration of breastfeeding from the prenatal to neonatal periods. Birth 2018;45(2):178–183. 10.1111/birt.12323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Kay MC, Cholera R, Flower KB, et al. Are Low-Income, Diverse Mothers Able to Meet Breastfeeding Intentions After 2 Months of Breastfeeding? Breastfeed Med 2020;15(7):435–442. 10.1089/bfm.2020.0025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Prevalence of Obesity Among Children and Adolescents: United States, trends 1963–1965 through 2007–2008. http://www.cdc.gov/nchs/data/hestat/obesity_child_07_08/obesity_child_07_08.htm. Accessed 2012.
- 43.Hernández-Cordero S, Pérez-Escamilla R. What will it take to increase breastfeeding? Maternal & Child Nutrition 2022;18(S3):e13371. 10.1111/mcn.13371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Bailey ZD, Krieger N, Agenor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. Lancet 2017;389(10077):1453–1463. 10.1016/S0140-6736(17)30569-X. [DOI] [PubMed] [Google Scholar]
- 45.Pope CJ, Mazmanian D. Breastfeeding and Postpartum Depression: An Overview and Methodological Recommendations for Future Research. Depress Res Treat 2016;2016:4765310. 10.1155/2016/4765310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Dias CC, Figueiredo B. Breastfeeding and depression: a systematic review of the literature. J Affect Disord 2015;171:142–154. 10.1016/j.jad.2014.09.022. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.