Breastfeeding offers numerous health benefits for birthing people (hereafter ‘mothers’) and children, yet these benefits are not equitably experienced by all mothers and children(1). The social determinants of health (SDoH) are potential modifiable factors that influence breastfeeding initiation and duration(Reference Standish and Parker2). Food and financial insecurity are complex SDoH factors that can potentially impact the ability of first-time mothers to breastfeed their infants. For example, food and/or financial insecurity may undermine access to nutritious and nourishing foods and/or force mothers to return to work shortly after giving birth, due to limited family leave policies at many US-based employers. Social welfare programs, such as the Special Supplemental Nutrition Program for Women, Infants and Children (WIC), may also play a significant role in influencing US mothers’ decisions to initiate and sustain breastfeeding. This program provides comprehensive nutrition and breastfeeding education to pregnant people and new mothers, which has the potential to encourage mothers to breastfeed. However, the provision of formula subsidies within WIC may inadvertently have the opposite effect by facilitating formula feeding instead(Reference Baumgartel and Spatz3). The burgeoning literature examining whether food insecurity and access to food assistance programs in the US influence breastfeeding duration has demonstrated inconsistent findings(Reference Gross, Mendelsohn and Fierman4–Reference Schultz, Byker Shanks and Houghtaling6). In addition, much less is known about the relationship between financial insecurity and breastfeeding duration.
The purpose of this study was to explore the relationships of food and financial insecurity, as well as food assistance, with breastfeeding continuation at four months postpartum in a sample of first-time mothers. Examining these relationships at four months postpartum is critical. This period represents a pivotal time in their breastfeeding journey, as mothers face many important decisions, including introducing solid foods and navigating challenges related to returning to work while maintaining breastfeeding(Reference Gianni, Bettinelli and Manfra7). We further explored whether these relationships differed by race and ethnicity given existing disparities in breastfeeding prevalence(Reference Thulier and Mercer8). Understanding these relationships will provide insights into whether interventions aimed at improving food and financial support could improve breastfeeding goals and reduce breastfeeding disparities, ultimately promoting equitable breastfeeding outcomes to improve the health of both mothers and infants.
Methods
Data are from the Life-course Experiences And Pregnancy (LEAP) study, a retrospective cohort of perinatal health among 977 women participating since adolescence in the Eating Activity in Teens and Young Adults (Project EAT), a longitudinal cohort study. Project EAT participants were recruited from the Minneapolis-St. Paul, Minnesota, public schools at ages 11–18 years from 1998–1999. Data were collected every five years, with the most recent EAT survey (EAT-IV) completed during 2014–2015. A flowchart (Fig. 1) details data collection, response rates and inclusion criteria. A detailed description of the Project EAT study aims, methods and main findings has been published elsewhere(Reference Neumark-Sztainer, Wall and Chen9–Reference Neumark-Sztainer, Story and Hannan11). The analytic sample of the current study was restricted to parous people who self-identified as women or girls on the baseline EAT-I survey and who provided complete responses on food and financial security during pregnancy and infant feeding practices. In addition, we restricted to those who indicated exclusively or mostly feeding either formula or breastmilk at four months postpartum to ensure distinct comparison groups (n 486; 74 % of 656). The demographic distribution from the LEAP cohort, Project EAT-I and the current LEAP subsample is similar. All study protocols were approved by the University of Minnesota’s Institutional Review Board Human Subjects Committee, and electronic consent was obtained from all study participants.
Fig. 1 Flowchart of Project EAT, LEAP and LEAP subsample participants
Measures
Survey items and coding decisions are presented in Table 1. The LEAP survey was designed to gather retrospective information from participants, specifically targeting non-overlapping timeframes. Specifically, food and financial insecurity and food assistance questions referred to during pregnancy with their first child period, and breastfeeding practices referred to four-month postpartum period. Infant feeding practices were assessed from a single adapted item indicating the relative proportion of breastmilk v. formula used for feeding within the first four months postpartum(Reference Shulman, D’Angelo and Harrison12,Reference Oken, Baccarelli and Gold13) . Participants who indicated using half formula/ half breastmilk (n 117) or did not provide an answer (n 1) were excluded from the analysis. This decision aimed to ensure the creation of distinct comparison groups, representing clear preferences for either breastfeeding or formula feeding. Food assistance, security and financial security were assessed using an adapted measure from Blumberg’s short form of the Household Food Security Scale(Reference Blumberg, Bialostosky and Hamilton14).
Table 1 Description of measures and coding decisions
Statistical analysis
In this secondary analysis of the LEAP study, we used modified Poisson regression models(Reference Zou15) with robust SE to explore the associations between food and financial insecurity and food assistance, with breastfeeding continuation at four months postpartum. These models estimated prevalence ratios and 95 % CI. We also explored effect measure modification by self-reported race and ethnicity by adding interaction terms to all models. Models were adjusted for age at first birth, relationship status during pregnancy and self-reported race and ethnicity to account for historical social construction of race and ethnicity categories. The interpretation of the results was focused on the magnitude, direction and precision (95 % CI) of effect estimates rather than relying on significant/non-significant interpretations.
Results
Baseline characteristics of participants
Nearly two-thirds of mothers self-identified as White and reported being married or in a domestic partnership (Table 2). Approximately two-thirds of mothers exclusively or mostly breastfed their infants during the first four months compared to one-third of mothers who exclusively or mostly formula-fed their infants.
Sociodemographic-adjusted associations of food insecurity, financial insecurity and food assistance during pregnancy with breastfeeding for the first 4 months postpartum
Table 2 Participant characteristics during the first pregnancy
Note: Numbers and % may not add up to 100 % due to missing observations.
Experiencing food insecurity during pregnancy was associated with 0·65 (95 % CI = 0·43, 0·98) times the prevalence of exclusively or mostly breastfeeding during the first four months postpartum compared to food security, after adjustments for sociodemographic covariates. Mothers who experienced financial insecurity had 0·92 (95 % CI = 0·78, 1·08) times the prevalence of exclusively or mostly breastfeeding during the first four months postpartum relative to mothers who were financially secure, after adjustments for sociodemographic factors. Receipt of food assistance, including accessing a food shelf or receiving benefits through safety net programs including WIC during pregnancy, was associated with 0·66 (95 % CI = 0·49, 0·88) times the prevalence of breastfeeding during the first four months postpartum, compared to no food assistance, after adjustments for sociodemographic covariates (Table 3).
Table 3 Associations of food and financial insecurity and food assistance with breastfeeding continuation at four months of age and effect measure modification by race and ethnicity
Model 2 was adjusted for maternal age, race/ethnicity and relationship status.
* Prevalence ratio.
Effect measure modification by self-reported race and ethnicity
We did not find evidence of effect measure modification by race and ethnicity (P values for all statistical interactions were >0·2).
Discussion
Our findings showed that mothers who reported experiencing food insecurity and those using food assistance programs during the prenatal period had a lower prevalence of breastfeeding at four months postpartum relative to mothers who did not report food insecurity or access to food assistance programs. Our results for financial assistance were close to the null. The point estimate for financial insecurity indicated a slightly lower prevalence of breastfeeding but 95 % CI were consistent with both a 22 percent lower likelihood of breastfeeding and an 8 percent higher likelihood of breastfeeding at four months after childbirth. Lastly, we did not find evidence for effect measure modification by race and ethnicity. These findings extend the literature documenting the impact of SDoH, such as food insecurity and access to food assistance programs on infant feeding practices.
Our findings of a negative association between food insecurity during pregnancy and breastfeeding are consistent with both observational and qualitative studies(Reference Orozco, Echeverria and Armah16,Reference Gross, Mendelsohn and Arana17) . The premature discontinuation of breastfeeding among mothers living in food-insecure households reduces the likelihood of both mothers and infants experiencing the health-related benefits linked to breastfeeding(Reference Kramer and Kakuma18,Reference Horta and Victora19) . For mothers, they are less likely to benefit from the added health gains that breastfeeding can promote, such as potential for a lower risk of developing hypertension, hyperlipidaemia, type 2 diabetes, different types of cancers (e.g. breast, ovarian, endometrial cancer) and CVD(Reference Sattari, Serwint and Levine20). Infants are more likely to be deprived of key nutrients offered by breast milk, which play a vital role in strengthening their immune system and safeguarding them against a range of illnesses in the long term, such as asthma and certain infections(Reference Ip, Chung and Raman21). Thus, addressing food insecurity, in addition to being important in and of itself, may support breastfeeding, promoting the health and well-being of both mothers and infants.
We also found a negative association between receiving food assistance during the prenatal period and breastfeeding at four months after childbirth. In the US, programs such as WIC provide nutrition and breastfeeding support and education to pregnant and postpartum people living in low-income households. While the WIC program has made notable strides in enhancing maternal and child health outcomes, observational studies have shown that mothers participating in WIC have a lower prevalence of breastfeeding continuation and are more likely to introduce formula earlier than current recommendations(Reference Tenfelde, Zielinski and Heidarisafa22,Reference Panzera, Castellanos-Brown and Paolicelli23) . While the decision to stop breastfeeding is complex and multi-factorial, we note that the availability of free formula through WIC could be one of the many factors influencing the observed greater likelihood of formula feed in this sample. In the context of the US, paid parental or family leave is not federally mandated. In practice, this means that many mothers return to work shortly after giving birth due to economic pressures and concerns for job security. Compounded with workplace conditions that may foster an unsupportive atmosphere for breastfeeding (e.g. lack of lactation rooms, supportive culture and flexible work schedules) and childcare providers who might lack familiarity with handling breastmilk, these factors collectively pose an obstacle or impede mothers’ ability to sustain breastfeeding. Given these structural barriers, along with the pervasive availability of formula in the market influenced by the lobbying efforts of the milk industry, it is not surprising that many mothers who are relying on food assistance programs like WIC are more likely to introduce formula early in their infant’s feeding journey(Reference Baker, Smith and Garde24). Thus, our results provide additional evidence for the need for future interventions targeting supports for longer breastfeeding duration and addressing the SDoH via policy change that provide and improve access to food and cash assistance programs, affordable and nutritious foods, federally mandated paid family leave and other resources to promote a conducive home and workplace environment for mothers who choose to breastfeed.
Lastly, we did not find evidence of an association between financial insecurity and breastfeeding. This finding is consistent with a prior observational study that investigated a number of indicators of socioeconomic status and reported that family income was not associated with breastfeeding(Reference Heck, Braveman and Cubbin25). More research is needed to further understand how various dimensions of financial insecurity, such as holding multiple jobs or facing unstable income, may impact breastfeeding practices. Nonetheless, while financial insecurity may not appear to be a direct barrier to breastfeeding in this sample, policies and programs aimed at reducing income inequality and increasing access to food assistance programs are important in supporting maternal and child health outcomes. To develop equitable breastfeeding promotion and support strategies for mothers, future research is needed to understand other potential barriers to breastfeeding (e.g. cultural, psychosocial, physiological and structural) in this population and elucidate the mechanisms through which food and financial insecurity and food assistance may adversely or positively affect breastfeeding initiation and duration.
Our findings should be interpreted in light of our study limitations. First, our study is limited by retrospective self-report measures, which are subject to recall and social desirability biases and could lead to measurement error. Second, our findings might not be generalizable to other geographical areas or to mothers of different cultural or socioeconomic backgrounds. More research is needed among racially, ethnically and socioeconomically diverse populations. Third, we only have data about infant feeding practices after the first live birth. We recommend future studies explore the consistency of our findings across subsequent live births.
Conclusions
We found a lower prevalence of breastfeeding among mothers that reported experiencing food insecurity and using food assistance programs during the prenatal period. Policy-level changes tackling structural and social determinants of breastfeeding initiation and duration, including enforcing federally mandated paid family leave, workplace support for breastfeeding and improving access to affordable and nutritious foods are needed to support longer breastfeeding duration among mothers and advance breastfeeding equity.
Acknowledgments
The authors thank the LEAP staff and study participants for their important contributions.
Financial support
This research was supported by grant number R01HD090053 from the National Institute of Child Health and Human Development (PI: SM Mason). JN de Brito’s research was supported by the National Institute of Arthritis Musculoskeletal and Skin Diseases of the National Institutes of Health under Award Number K12 AR084223 (PI: S Allen). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funders had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Conflict of interest
None.
Authorship
J.N.d.B. and J.K.F. conceptualised the study, performed all data analysis, interpreted the data and wrote the original draf of the manuscript. S.M.M. secured funding and is the principal investigator of the LEAP study. S.T.J., J.M.B. and S.M.M. critically revised the manuscript and contributed to data interpretation. All authors read and approved the final manuscript.
Ethics of human subject participation
This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving research study participants were approved by the Institutional Review Board of the University of Minnesota. Electronic informed consent was obtained from all subjects.
Open access
