Abstract

In 2009, a pediatric hospital chaplain, distressed by signs of hunger among her patients’ caregivers, shared her concerns with physician leaders of an effort to improve local urban health conditions. She related her observations of hungry parents asking nurses, residents, and medical students for food. She witnessed the kindness of many providers who shared their own sandwich or a few dollars. She also described what she perceived to be a loss of dignity on the part of the person asking for food and the painful resignation of those who were unable to help. She worried that caregiver hunger could be an impediment to a child’s recovery.

Food insecurity is a major public health problem in the United States. One in 5 US households with children is food insecure.1 Household food security is defined as having “access at all times to enough food for an active, healthy life for all household members.”1^(p4),2 Acute symptoms of household food insecurity include hunger, depression, fatigue, and lack of control.3–6 Furthermore, people with household food insecurity can experience “alienation” and related feelings including “embarrassment and shame” and a sense of “powerlessness.”3 A qualitative study reported that adults and children with household food insecurity have difficulty concentrating and an impaired ability to learn.4 Hungry children are more likely to suffer from psychosocial problems, receive special education services, and fail a grade.7

The long-term outcomes of food insecurity include increased risk of poor physical and mental health in children8–11 and adults.12,13 Food insecurity is also associated with poor disease management and medication nonadherence.12,14,15 The prevalence of household food insecurity has been assessed in inpatient and outpatient populations; the prevalence in these populations frequently exceeds national population-based averages.16–20

Although household food insecurity is a known risk factor for poor health, chronic disease, and hospitalization in children,10,11,21,22 very little is known about the patterns and prevalence of food insecurity in the hospital setting. Results from a single qualitative study of 24 caregivers of hospitalized children found that among the personal, financial, and social costs of a child’s hospitalization, “parents experienced difficulty providing themselves with adequate and affordable food and drink. . . .”23^(p748) We found no published studies examining how the symptoms of food insecurity might be associated with a caregiver’s ability to play an active role in the care of a hospitalized child. Little evidence was found to inform a rapid, sustainable remedy to the problem of food insecurity among caregivers of hospitalized children.

At the time the chaplain approached us, the hospital’s policy permitted social workers to provide 1 free food tray per stay to families in need, regardless of the length of the patient’s stay. The chaplain’s concerns were validated by other professionals with experience as providers, caregivers, or both in the children’s hospital. Given the acuity of the situation, we opted for a “feed first, ask questions later” approach in alignment with the ethical principle of beneficence in research24 and the principles of our community-engaged research model.25 We marshaled medical students, other faculty, hospital personnel, and existing relationships with the Greater Chicago Food Depository to deploy the area’s first hospital-based food pantry.

After we implemented the food pantry, we launched a study to test the hypothesis that food insecurity among caregivers of hospitalized pediatric patients (“hospital food insecurity”) would be prevalent and associated with potentially mutable barriers to food access. We used these empirical findings to develop a conceptual framework to facilitate future study of the impact of hospital food insecurity on pediatric health and health care outcomes.

METHODS

This study was conducted at a 155-bed academic children’s hospital in Chicago, Illinois. The hospital admitted approximately 5000 patients annually; an estimated 60% were Medicaid beneficiaries. The hospital’s dining options included commercial cafés and a hospital-run cafeteria. A meal (e.g., sandwich, fruit, and drink) cost $10 to $12 per person. Few low-cost, nutritious alternatives were available within walking distance.

RESULTS

Of 234 eligible caregivers approached for this study between June and December 2011, 200 participated (cooperation rate 85%). No additional data on those who refused or who were not approached for participation were available. We excluded 8 participants from this analysis because they did not answer the hospital food insecurity survey item. Table 1 summarizes sociodemographic characteristics of caregivers. The vast majority of caregivers identified themselves to survey staff as a parent of the hospitalized child (relationship to the child was not ascertained on the survey).

A third of caregivers screened positive for hospital food insecurity; 66% of these caregivers also met criteria for household food insecurity in the previous 12 months (Figure 1). Odds of hospital food insecurity were significantly higher among caregivers who were Black or African American, single or divorced, less educated, had household food insecurity, received government food assistance (e.g., food stamps, Supplemental Nutrition Assistance Program), and who received a food donation in the previous 12 months. The odds of experiencing hospital food insecurity remained significantly higher for those identifying as Black or African American compared with Whites (adjusted odds ratio = 2.7; 95% confidence interval = 1.1, 6.6) when we controlled for receipt of government food assistance.

Open in a separate window

FIGURE 1—

Prevalence of food insecurity in caregivers of hospitalized pediatric patients: Chicago, IL, June–December 2011.

Note. Percentages do not total 100 because of rounding.

Financial barriers to food access were prevalent: 43% of all caregivers reported not having enough money to buy food in the hospital. These caregivers, as well as those who did not have reliable transportation, were significantly more likely to have experienced hospital food insecurity. Most caregivers reported that they were unlikely to leave the hospital during their child’s stay (69%) and nearly half felt that their child would not be OK if the caregiver left the hospital (48%). These barriers were not significantly associated with hospital food insecurity in bivariate analyses (Table 1).

In multivariable analysis (Table 2), odds of experiencing hospital food insecurity were significantly elevated for those who thought their child would not be OK if the caregiver left the hospital. Other factors independently associated with a higher odds of hospital food insecurity included household food insecurity in the previous 12 months, receipt of government food assistance in the previous 12 months, not having enough money for food in the hospital, and lacking reliable transportation.

DISCUSSION

To our knowledge, this is the first study to describe the prevalence of hospital food insecurity among caregivers in the pediatric hospital setting. At this urban academic children’s hospital, a third of caregivers met criteria for hospital food insecurity. More than 40% of caregivers in this study met criteria for household food insecurity in the previous 12 months, a prevalence similar to that reported for US households with children with income below the poverty level (45%) and households with children headed by single women (35%).1

Hospital food insecurity was associated with sociodemographic characteristics as well as with mutable factors, including having enough money to buy food at the hospital, having transportation to get food while at the hospital, and knowing where to get food while at the hospital. Although there were no previous studies of hospital food insecurity to corroborate these findings, the associations we found between household food insecurity and sociodemographic characteristics mirror associations reported for household food insecurity, including higher prevalence among people who are Black or African American, single or divorced, or less educated.1,12 Low income status could explain these sociodemographic associations with hospital food insecurity. When we used government food assistance as a proxy to adjust for income status, African American or Black race was independently associated with hospital food insecurity. This finding suggests, as has been found in previous studies of the relationship between race and other health outcomes,27,28 that factors more common to the experience of being African American or Black, including discrimination or racism, could be contributing to these racial differences in hospital food insecurity.

Previous studies found food insecurity in adults to be associated with symptoms including hunger, depression, fatigue, lack of concentration, lack of control, and impaired learning.3–6 Although beyond the scope of our study, these kinds of symptoms could impair caregiver decision-making and be misunderstood by health care providers as disinterest in, distrust of, or lack of cooperation with the child’s medical care. We found it interesting that a caregiver’s worry that the child would not be OK if left alone at the hospital was independently associated with hospital food insecurity. Whether worry about the child’s well-being prevents the caregiver from leaving to find food, or hunger promotes feelings of worry or distrust in the hospital care, cannot be discerned from this cross-sectional observation. However, it seems plausible that interventions to prevent and alleviate caregiver hunger in the hospital setting could promote trust between caregivers and staff. A prospective study is needed to assess the impact of the hospital food pantry or other hunger-prevention interventions on pediatric, caregiver, and even health provider outcomes.

The path to most health care interventions begins with research; however, the urgency of need and our capability to quickly respond warranted the “feed first, ask questions later” approach. Operating through volunteers and purchasing food with modest grant and philanthropic funds, our hospital-based pantry served approximately 10% of families of all children hospitalized during the study period (about a third of those in need, according to the findings of this study). Recipients expressed appreciation; 1 person called the program a “blessing” and another said it enabled him or her to “get through” the child’s hospital stay. In 1 instance, during a child’s hospitalization for 20 days of radiation therapy, an anonymous parent contributed donations supplied by supportive friends and neighbors and posted a letter encouraging other families to “pay it forward” (Figure 2).

Open in a separate window

FIGURE 2—

Note received from caregiver contributing to the food pantry at a Chicago, IL, pediatric hospital.

Nearly 40% of caregivers were receiving government food assistance suggesting that, in addition to, or as an alternative to, a food pantry, hospitals could work with their food vendors to provide items eligible for food stamp purchase. Since we began this work, we have identified 2 other hospital-based pantries, including the pediatric clinic at Hennepin County Medical Center29 and the Preventive Food Pantry at Boston Medical Center.30 Unlike our pantry, which focuses on feeding caregivers while their child is hospitalized, these pantries provide households with food to take home. Ideally, a hospital-based food pantry would offer both services.

On the basis of this study’s findings, the published literature, and expert opinion, we propose a preliminary conceptual framework to facilitate future research and the design of effective interventions to address hospital food insecurity (Figure 3). This framework highlights gaps in the current evidence base and informs future intervention by showing that a holistic intervention would connect caregivers to food in the hospital and then extend into the home (more than 40% of children were discharged to food-insecure homes) to prevent caregiver and child food insecurity after a child’s hospitalization.

Open in a separate window

FIGURE 3—

Preliminary conceptual framework for risks for, and the acute and long-term outcomes of, hospital food insecurity in caregivers of hospitalized patients derived from literature, study results, and expert opinion: Chicago, IL, June–December 2011.

Note. FI = food insecurity.

^aOutcomes supported by expert opinion and clinical experience; additional research needed.

Acknowledgments

Human Participant Protection

References