Food Insecurity During Pregnancy and Breastfeeding by Low-Income Hispanic Mothers

We performed a qualitative study of low-income Hispanic mother-infant pairs participating in an early child obesity prevention study at a large urban public hospital in New York City. This study was approved by the institutional review boards of the New York University School of Medicine, Albert Einstein College of Medicine, Bellevue Hospital Center, and New York City Health and Hospitals Corporation and was registered on www.clinicaltrials.gov (NCT01541761).

This qualitative study was nested within a randomized controlled trial testing the efficacy of a primary care–based early child obesity prevention program, called Starting Early, designed for low-income Hispanic mother-infant pairs, beginning in pregnancy and continuing until child age 3 years.²⁰ The main components were (1) individual prenatal and postpartum nutrition counseling and (2) nutrition and parenting support groups coordinated with well-child visits. The intervention was not designed to directly address food insecurity. For the larger trial, we enrolled pregnant women in their third trimester who were at least 18 years old, self-identified as Hispanic and/or Latina, were fluent in English or Spanish with a singleton uncomplicated pregnancy, and who intended to receive care at the study sites. We excluded women with significant medical or psychiatric illness, homelessness, or severe fetal anomalies.

We employed purposive sampling of mothers randomly assigned to the Starting Early Program intervention to obtain maximum variability across 3 developmental stages of feeding: (1) early infancy (3–7 months old), (2) late infancy (10–15 months old), and (3) toddlerhood (19–24 months old). Early infancy reflected the period of predominate milk feeding. Late infancy reflected increased consumption of solid foods and independent self-feeding. Toddlerhood reflected increased independence. Given that food insecurity is pervasive in the context of poverty and that even marginal food insecurity is associated with decreased health,²¹ we interviewed mothers with negative, marginal, and positive food insecurity screens to identify family strengths and vulnerabilities. Trained staff contacted eligible mothers between March 2015 and October 2015 and obtained written informed consent for the 1-time qualitative interview.

Interview guides (Table 1) were informed by the Core Food Security Module from the US Department of Agriculture,²² a scale used nationally to assess the extent and severity of food insecurity. The interview guide was designed by a multicultural interdisciplinary team of general pediatricians, a developmental and behavioral pediatrician, a medical student, and registered dietitians. Trained, bilingual English-Spanish interviewers conducted the interviews, ranging from 40 to 70 minutes. Interviews were audio recorded, professionally transcribed and translated, and reviewed by staff to verify translation accuracy.

Four researchers, consisting of a general pediatrician, a registered dietitian, a medical student, and a research assistant, used an iterative process of textual analysis. The group used the constant comparative method to identify and iteratively refine codes and devise a final code structure.²³ The group initially read the transcripts in their entirety to gain insight into broad themes, discussed individual thoughts and repeating themes, and confirmed consistency, coherence, and distinctiveness before finalizing the codebook. Consensus was reached regarding each quotation. Data organization, retrieval, and stratification by child age was facilitated by Dedoose software. Interviews were conducted until thematic saturation was reached.

Sociodemographic data collected from the trial included maternal age, parity, education, work, marital status, country of origin, and participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and the Supplemental Nutrition Assistance Program (SNAP). Food insecurity was determined by using the US Department of Agriculture Core Food Security Module.²² Continuous scores were generated from 10 questions assessing food security in the previous 12 months. Scores were categorized by using recommended cut points (food secure, 0; marginally food insecure, 1–2; and food insecure, 3–10). Difficulty paying bills in the previous 12 months was assessed by using 2 questions from the Survey of Income and Program Participation and was defined as either experiencing difficulty paying monthly bills, rent, or mortgage and/or having a service turned off because payments were not made.²⁴ 

We sampled 100 mothers who were primarily born outside the United States (87%), Spanish speaking (87%), WIC participants (91%) with infants (32 with 3–7-month-olds, 31 with 10–15-month-olds, and 37 with 19–24-month-olds). The majority reported experiencing food insecurity (67%; Table 2).

Three main themes, each with subthemes, are described below. Example quotations are organized by theme and numbered consecutively in tables.

See Table 3 for quotations related to contributors. Difficulty meeting basic needs was a common strain. Rent, phone, and electricity bills were hard to manage simultaneously (quotation 1). Paying for household expenses was often prioritized over buying food (quotation 2).

Job instability, prolonged unemployment, and loss of work hours led to increased vulnerability to food insecurity. When fathers were unemployed, families had fewer food choices and ate on the basis of availability. Families became accustomed to having certain income, so the greatest hardship occurred immediately after a job loss (quotation 3). One mother described job loss as a “crisis” that can force families to purchase only necessities (quotation 4).

Pregnancy and infancy increased vulnerability to food insecurity. During pregnancy, mothers described strain due to changes in employment attributed to frequent medical visits (quotation 5), physical limitations, and health complications (quotation 6). New infant expenses forced families to choose between buying diapers and paying bills (quotation 7), leading to feelings of desperation (quotation 8).

Financial strain unique to immigration included the concept of “double expenses” for families with children in both the United States and their country of origin (quotation 9). Mothers who “support 2 homes” discussed difficulties balancing their own needs with familial obligations in their native country stemming from needing to send money for their children’s housing and food (quotation 10).

See Table 4 for quotations related to feeding. Common concerns among mothers of young infants related to breastfeeding. Mothers may avoid breastfeeding because of concerns about their own diet. Mothers perceived their diets as poor and worried that if they did not eat enough fruit and vegetables, their breast milk would be of low quality and lack the necessary nutrients (quotations 11 and 12).

Mothers experienced substantial stress related to food insecurity and difficulty paying bills. Some mothers felt stress would diminish the quantity of their breast milk (quotation 13) or cause the milk to “dry up” (quotation 14). Mothers worried that if breast milk was insufficient, formula would be needed, which was expensive and not always available.

Mothers tried to hide their worries during feeding, fearing that stress would diminish the infant’s appetite (quotation 15) or affect the infant’s mood (quotation 16). Mothers believed negative emotions could pass to the infant during feeding, either through physical closeness or through the breast milk itself. One mother recalled “elders” cautioning to avoid breastfeeding when stressed because feelings transmit through the breast milk (quotation 17) and that milk may need to be discarded (quotation 18). The belief that infants could absorb the mother’s poor energy during feeding led to family members discouraging breastfeeding and encouraging formula (quotation 19).

Mothers of older infants and toddlers described actively limiting food. Mothers limited each family member’s portion sizes (eg, by giving “just 1 piece of meat”; quotation 20). Healthier, more expensive food, such as fruits and vegetables, were limited by “not putting too many on the table” or substituting less expensive foods (quotations 21 and 22). Active limiting was a tool used to prevent the family from not having enough food the next day (quotation 23).

See Table 5 for quotations related to home- and community-level coping strategies. Home strategies involved considerable planning. Families made shopping lists to prevent unnecessary purchases and food waste (quotation 24), avoided street foods, and cooked at home (quotation 25). Families limited other spending, such as electricity, by turning off televisions and lights (quotation 26) and delayed spending on other bills. Many expressed that “food must be first” and delayed buying clothes and shoes (quotation 27) or paying less important bills, such as the telephone (quotations 28 and 29). Mothers relied on inexpensive staples that stored easily and would not spoil, such as beans, rice, and tortillas (quotations 30–32).

Social support networks, such as extended family (quotations 33–35), friends (quotation 36), and churches (quotation 37), provided safety nets when household strategies proved insufficient. Networks either loaned money or directly provided food. Living with extended family allowed for lower rent burden, decreased expenses, and the distribution of bills among family members.

Supplemental nutrition assistance programs, specifically WIC and SNAP, provided critical assistance, although some expressed eligibility barriers. Some felt that without these services, their food insecurity would be unmanageable (quotations 38 and 39). Other mothers noted difficulty obtaining services (quotation 40) or the funds not lasting the whole month (quotation 41).

In this inner-city, mostly immigrant, low-income, Hispanic sample, we found that mothers frequently experienced food insecurity during pregnancy, infancy, and toddlerhood. Contributors included difficulty meeting basic needs, job instability, and vulnerabilities specific to pregnancy, infancy, and immigration. Mothers perceived food insecurity to affect feeding practices. In early infancy, women worried that poor maternal diet and stress due to food insecurity adversely affected breast milk quantity and quality, leading to perceived need to supplement with formula or discard breast milk. In late infancy and toddlerhood, mothers described actively limiting expensive healthy foods, including fruits, vegetables, and lean meats. Coping strategies within the home and those relying on outside support mitigated food insecurity. These findings can inform recommendations to assist families with healthy infant feeding within the context of food insecurity and poverty more broadly.

Food insecurity is dynamic and complex. Changes in income, expenses, or access to government assistance may diminish abilities to purchase healthy foods.²⁵ Our findings are consistent with studies of broader populations that have described similar experiences with shifting household composition, job instability, housing instability, and unexpected expenses as contributors.^26,–29 Hispanic families, particularly immigrant families, have unique vulnerabilities that shape their experience and ability to cope.³⁰ Immigrant families had to choose between feeding their households or sending money to their families, often in more dire situations, in their home countries.²⁸ Remittances are often prioritized when children are left behind.³¹ 

Contributors unique to pregnancy and infancy included decreased employment and increased expenses, including diapers, clothes, and formula. Food insecurity during pregnancy has been associated with high gestational weight gain, gestational diabetes mellitus, disordered eating, and dietary fat intake postpartum^16,32 as well as higher prenatal stress and obesity-promoting infant-feeding attitudes.^16,33 Food insecurity before and after birth has been associated with nonresponsive maternal infant-feeding styles and obesity-promoting practices.^12,13,15 A key contribution of this study has been the identification of several potential mechanisms linking food insecurity to maladaptive infant-feeding beliefs, styles, and practices, specifically limiting breastfeeding and more expensive healthy foods. These potential mechanisms are each discussed below.

During early infancy, mothers described limiting breastfeeding because of fears that their own poor diet would adversely affect breast milk quantity and quality. Evidence about how maternal diets that are low in macro- and micronutrients may influence breast milk composition has been mixed.³⁴ Some studies have linked nutrient levels consumed during lactation, such as vitamin C and fish oils, to their concentration in human milk, particularly in low-income Latino countries.³⁵ However, others found that overall adequacy of human milk is maintained regardless of maternal nutrition.³⁶ Providers should address these misperceptions and reassure women that exclusive breastfeeding remains the optimal feeding modality and provides health benefits for mothers. Counseling about these concerns can provide the opportunity to address maternal nutrition in the context of food insecurity, including access to supplemental nutrition-assistance programs and the need to continue prenatal vitamins during lactation. Unnecessary formula supplementation due to these misperceptions may cause overfeeding and represent 1 pathway linking food insecurity during pregnancy and infancy to child obesity.

Although maternal concern about the adverse effects of stress on breastfeeding has been documented,^37,38 no studies, to our knowledge, have explored the belief that maternal stress can be transmitted through breast milk itself. Pregnancy-related stress, generalized anxiety, and postpartum depression have been found to decrease initiation and continued breastfeeding.³⁹ Studies of low-income women found that stressful life events, in particular financial stressors, were associated with shorter breastfeeding duration. Although there is limited evidence, the existing studies do not suggest that breast milk composition is changed by negative emotions.⁴⁰ The decreased initiation, exclusivity, and duration of breastfeeding found in low-income families may be related to this belief, which may represent another link between food insecurity and child obesity. Providers should explore maternal concerns about how stress impacts breastfeeding, provide reassurance that breast milk does not transmit maternal negative emotions, and offer mental health services to enhance stress management.

We found that when food supplies were low, mothers relied on inexpensive staples or actively limited variety, consistent with studies of families with older children.²⁶ Food insecurity has been associated with higher intake of food staples such as beans and tortillas among preschoolers, which are likely bought in bulk to use when families cannot afford perishable items.⁴¹ Limiting expensive and perishable foods, such as fruits and vegetables, may hinder opportunities to introduce these foods to young children, negatively affecting development of food preferences. Restrictive feeding has been associated with poor child self-regulation during feeding and increased inappropriate weight gain.^42,43 Restricted access to less nutritious snack foods during periods of food insecurity may increase child demands for these foods during times of relative food security.⁴¹ 

Positive coping strategies to secure money for food occur at multiple levels. Within the home, proactive rather than reactionary coping strategies were more effective in reducing or avoiding food insecurity.²⁶ Protective factors or strategies that are amenable to change included planning ahead for necessary expenses, only buying necessities, relying on staples, and delaying spending. When these strategies proved insufficient, families relied on social support networks. Those lacking family support depended on community and religious organizations. Perceptions toward government assistance programs varied, with some families finding relief and others reporting difficulties receiving enough. Federal benefits attenuate the severity of food insecurity but might not eliminate it, particularly for children and in regions with higher food costs.²⁵ Health care providers should advocate for increased support for these programs, ensure that families are aware of their eligibility, and provide assistance obtaining these resources.

To date, the American Academy of Pediatrics²⁵ has published recommendations to incorporate screening for food insecurity in primary care and connecting at-risk families to services that provide food directly or enhance enrollment into assistance programs. However, these guidelines do not include how families could optimize nutrition in the context of scarce resources. Separate guidelines published by the National Academy of Medicine (formerly the Institute of Medicine) and the American Academy of Pediatrics regarding the prevention of child obesity address promoting healthy infant diets and responsive feeding styles.^44,45 However, these guidelines do not incorporate the need for food insecurity screening to address barriers to implementing infant-feeding recommendations. In the future, policy makers should consider combining food insecurity and child obesity prevention recommendations to enhance the care of families living in poverty.

Limited primary care interventions exist to improve infant feeding in the context of food insecurity. Group prenatal care interventions have been found to improve budget-management skills, diet, and self-efficacy in affording healthy foods.⁴⁶ The Keeping Infants Nourished and Developing program, a collaborative partnership between pediatric primary care and a community food bank, found that participants were more likely to connect with social work or the medical-legal partnership to reduce food insecurity.⁴⁷ Although these programs focus on reducing food insecurity, they do not specifically address infant-feeding practices. The Starting Early Program, a primary care–based early obesity prevention intervention,²⁰ exemplifies an infrastructure that could directly link families to community services during pregnancy and infancy and simultaneously promote healthy infant-feeding practices. Future adaptations will explore the feasibility and efficacy of these modifications.

Our study has several limitations. Given that the sample included inner-city, mostly immigrant, low-income, Hispanic mothers, generalizability may be limited. Additionally, we may not have included individuals with the fewest coping strategies for managing food insecurity given that the trial excluded the most marginalized families, including those with homelessness or severe mental illness. In future studies, researchers should aim to better understand quantitatively the themes identified in this report by studying longitudinal associations between food insecurity, maternal diet, maternal stress, breastfeeding, limiting of healthy foods, and later, child obesity.

In this study, low-income Hispanic mothers with infants and toddlers perceived food insecurity to adversely affect breastfeeding and to require actively limiting healthy foods and portion sizes. These maternal beliefs and behaviors may represent potential mechanisms linking food insecurity and obesity-promoting feeding styles and practices. Although the single most important action would be to reduce food insecurity during these sensitive periods, stakeholders in programs and policies to prevent poverty-related disparities in child obesity should consider and address the broader context by which food insecurity is associated with contributing beliefs, styles, and practices.

We thank the Starting Early Program staff who contributed to this project.

SNAP

Supplemental Nutrition Assistance Program

WIC

Special Supplemental Nutrition Program for Women, Infants, and Children