Food Insecurity and Community-Based Food Resources Among Caregivers of Hospitalized Children

Abstract

Objective: Children's hospitals are implementing interventions to connect families to community-based resources. This study describes food insecurity (FI) and food resource knowledge, need, and use among families with a hospitalized child.

Methods: Between November 2020 and June 2022, 637 caregivers of hospitalized children in an urban 42-ZIP-code area were surveyed as part of a randomized controlled trial. The United States Department of Agriculture 18-item Household Food Security Survey was used to evaluate 12-month food security (food secure [score of 0=FS]; marginally secure [1-2=MFS]; insecure [3-18=FI]). Food resource knowledge, need, and use were described by food security status and examined using Cochran-Armitage tests. The distribution of local resources was obtained from a database and mapped by ZIP code.

Results: Comparing FI (35.0%) with MFS (17.6%) and FS (47.4%) groups, the rates of resource knowledge were lower (70.2% vs 78.5%, 80.5%), and the rates of need (55.1% vs 30.6%, 14.2%) and use (55.3% vs 51.4%, 40.8%) were higher. Rates of food resource knowledge increased linearly with increasing food security (FI to MFS to FS; P = .008), whereas the rates of resource need (P < .001) and use (P = .001) decreased with increasing food security. There were 311 community-based organizations across 36 ZIP codes with participants (range/ZIP code = 0-20, median = 8).

Conclusions: Half of families with a hospitalized child experienced FI or MFS. Although families exhibited high food resource knowledge, nearly half of families with FI had unmet food needs or had never used resources.

FIGURE 1

Recruitment and enrollment Consolidated Standards of Reporting Trials diagram for the RCT from which the data for the current study were obtained. Not approached: research assistant was unable to make contact with the caregiver to assess interest in screening for inclusion. Includes caregiver not in room when recruiting in-person, no answer to the bedside or cell phone, and no response to recruitment text message. This designation was also chosen if the patient was discharged before contact could be made. No intent to approach: child patient was identified from the electronic medical record database and added to patient list, but caregiver was not approached because of (1) enrollment cap reached, (2) full enrollment completed, (3) physician opted out for this patient, or (4) conflict of interest. Not screened: caregiver expressed interest in study participation but was not screened either due to our inability to reach the caregiver for eligibility screening or the patient was discharged before we could screen them for eligibility. Declined to screen: caregiver was approached for inclusion in the study but declined to be screened for eligibility. Ineligible because of enrollment caps based on FI status: for a 2-week period during enrollment, research team enrolled only caregivers from households classified as FI. During this time, caregivers from FS households were deemed ineligible. Unable to consent: caregiver was unable to provide consent because of technical difficulties with the e-consent process. Declined to consent: caregiver completed eligibility screening but declined to proceed through the informed consent process. Not consented: caregiver did not complete the consent process because of our inability to reach the caregiver for consent, the patient was discharged before we could proceed through the informed consent process, or we reached our target enrollment during the screening process. Consented and declined: caregiver proceeded through the informed consent process but declined to participate. Discontinued: caregiver provided informed consent to participate in the study but did not complete enrollment before the patient was discharged.

FIGURE 2

Assessment of knowledge, need, and use regarding community-based food resources in the CommunityRx-Hunger baseline survey. Solid lines with filled arrows outline the flow of question and response items. Dotted lines with open arrows indicate the categories to which participants were assigned on the basis of their responses. Participants who responded “Don’t Know/Not Sure” to any of the above question items were not assigned to the corresponding categories.

FIGURE 3

Caregivers’ self-reported food resource knowledge, need and use, stratified by food security status. * P_trend <.05, ** P_trend <.01, *** P_trend <.001.

FIGURE 4

Community-based food resources in the study region by ZIP code. The distribution of free or low-cost food, free meal, and free or low-cost food delivery resources in ZIP codes having ≥1 study participant was obtained from an online resource referral database and mapped by ZIP code. The lightest grey-shaded ZIP codes include ZIP codes outside the study geography and other ZIP codes that were included in the study geography but in which no participants were enrolled (ZIP codes 60805, 60130, 60153, 60162, 60202, and 60302). The black area in the state legend indicates the region being shown in the 4 maps to the left.