Abstract
This cohort study assesses the association of inpatient food insecurity with length of stay and readmission among caregivers during hospitalization of their children.
Household food insecurity (HFI) is associated with poor health outcomes and increased health care use but is transient and dependent on context and environmental stressors, including hospitalization. Inpatient food insecurity (IFI) is a phenomenon identified as food insecurity (FI) experienced by caregivers during their child’s hospitalization, with or without HFI. Stressors associated with pediatric hospitalization include personal, financial, and social difficulties experienced by caregivers. Prior evaluation of IFI’s consequences revealed caregiver depression, fatigue, and potential distrust of health care practitioners. However, little is known about the implications of IFI for hospital outcomes. In this study, we assessed the association of IFI with length of stay (LOS) and readmission.
Methods
Data were prospectively captured at a children’s hospital within an academic medical center in North Carolina between May 2022 and December 2023. The Wake Forest University School of Medicine Institutional Review Board approved the study and waived informed consent because deidentified data were used. We followed the STROBE reporting guideline.
We measured HFI via the 2-item Hunger Vital Sign. We developed a 3-item IFI screening tool to identify whether the caregiver (1) ate less than they felt they should, (2) felt hungry but did not eat, or (3) skipped meals during their child’s hospitalization due to lack of money for food. This tool was adapted from modified household food security surveys created to assess IFI. Positive screen was defined as a yes response to at least 1 of the 3 items. Nursing staff were encouraged to screen caregivers verbally at least once during the hospitalization and to document results in the electronic health record. Caregivers with IFI were provided with food during the admission.
Participant characteristics were compared by IFI status using χ2 or unpaired, 2-tailed t tests. Multivariable regression was used to assess the association of IFI with LOS (linear regression) and readmission within 30 days (logistic regression), adjusted for patient age, sex, race and ethnicity, language, and insurance. Race and ethnicity were used as a proxy for experiences with structural racism. Analyses were performed using Stata SE 14.2 (StataCorp LLC).
Results
Of 9325 hospital encounters (4224 females [45%], 5101 males [55%]; median [range] age, 7.1 [7.0-7.3] years), 718 (8%) screened positive for IFI and 560 (6%) for HFI. Caregivers with IFI included 197 Black (27%), 163 Hispanic (23%), and 312 White (44%) individuals, and 208 individuals with other (29%) race and ethnicity; preferred English language (585 [82%]); and had Medicaid (625 [89%]) or no insurance (11 [2%]) (Table 1). Forty percent of those with IFI screened negative for HFI. Participants with IFI had a longer LOS (β coefficient, 2.41; 95% CI, 1.63-3.19 days) and higher odds of readmission within 30 days (odds ratio, 1.77; 95% CI, 1.31-2.38) vs those without IFI (Table 2).
Table 1. Baseline Characteristics of Patient Encounters Screened for Inpatient Food Insecurity.
| Characteristic | No. (%) | ||
|---|---|---|---|
| All patients screened for IFI (n = 9325)a | Without IFI (n = 8607) | With IFI (n = 718) | |
| Age, median (range), y | 7.1 (7.0-7.3) | 7.2 (7.1-7.3) | 6.4 (6.0-6.8) |
| Sex | |||
| Female | 4224 (45) | 3884 (45) | 340 (47) |
| Male | 5088 (55) | 4711 (55) | 377 (53) |
| Race and ethnicityb | |||
| Black | 2115 (23) | 1918 (22) | 197 (27) |
| Hispanic | 1528 (17) | 1365 (16) | 163 (23) |
| White | 5085 (55) | 4773 (56) | 312 (44) |
| Otherc | 2113 (23) | 1905 (22) | 208 (29) |
| Preferred language | |||
| English | 8422 (91) | 7837 (91) | 585 (82) |
| Spanish | 724 (8) | 609 (7) | 115 (16) |
| Otherd | 155 (2) | 141 (2) | 14 (2) |
| Insurance | |||
| Medicaid | 6278 (68) | 5653 (67) | 625 (89) |
| Commercial | 2720 (30) | 2657 (31) | 63 (9) |
| Other public | 140 (2) | 133(2) | 7 (1) |
| None | 35 (0) | 24 (1) | 11 (2) |
| HFI | 241 (6) | 45 (1) | 196 (60) |
Abbreviations: HFI, household food insecurity; IFI, inpatient food insecurity.
Of the 13 082 hospital encounters (for 9196 unique participants), 9325 included screening for IFI and 3514 included screening for HFI; 2945 did not have any documented screening.
Race and ethnicity were extracted from the electronic health record.
Other races and ethnicities include American Indian or Alaska Native; Asian Indian; Chinese; Filipino; Japanese; Korean; Latin American or Hispanic; Native Hawaiian; other; other Asian; Other Pacific Islander; Samoan; Vietnamese; patient refused; patient unable to respond; and unknown.
Other languages include Arabic, Burmese, Chinese (Mandarin), French, Haitian Creole, Hindi, Hmong, Karenni, Nepali, Other, Portuguese, Russian, Sign Language, Swahili, Tigrinya, Urdu, and Vietnamese.
Table 2. Association of Inpatient Food Insecurity With Hospital Outcomes.
| Outcome | Unadjusted models | Adjusted modelsa |
|---|---|---|
| Hospital length of stay, β coefficient (95% CI) | 2.61 (1.85-3.37) | 2.41 (1.63-3.19) |
| Odds of readmission within 30 d, odds ratio (95% CI) | 1.52 (1.15-2.03) | 1.77 (1.31-2.38) |
Models adjusted for patient age, sex, race and ethnicity, language, and insurance.
Discussion
As 40% of caregivers with IFI did not have HFI, screening for HFI alone may miss many families experiencing FI during their child’s hospitalization. Both IFI and HFI are distinct social risks; screening for IFI is warranted to provide appropriate nutritional and social support to caregivers, promote better health, improve hospital outcomes, and minimize disparities.
Study limitations include data from a single institution and lack of universal screening. The study occurred during the COVID-19 pandemic, when FI rates were lower than before and after the pandemic, likely due to temporary pandemic-related emergency food relief programs. Further research should evaluate the implications of IFI for child health outcomes and etiologies for increased LOS and readmission rate. For caregivers experiencing IFI, increased LOS and odds of readmission likely precede challenging social circumstances associated with hospitalization, such as foregone work time and transportation difficulties.
Developing interventions to reduce IFI may benefit hospitalized children and their families as well as hospital systems. Future work should explore health outcomes and quantify the potential economic consequences.
Data Sharing Statement
References
- 1.Lee AM, Bocchini C, Lopez M, Clegg K, Greeley C. Food insecurity in the caregivers of hospitalized pediatric patients. Pediatrics. 2019;144(2_meetingabstract):481. doi: 10.1542/peds.144.2MA5.481 [DOI] [Google Scholar]
- 2.Berkowitz SA, Terranova J, Randall L, Cranston K, Waters DB, Hsu J. Association between receipt of a medically tailored meal program and health care use. JAMA Intern Med. 2019;179(6):786-793. doi: 10.1001/jamainternmed.2019.0198 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Markowitz MA, Tiyyagura G, Quallen K, Rosenberg J. Food insecurity screening and intervention in United States children’s hospitals. Hosp Pediatr. 2022;12(10):849-857. doi: 10.1542/hpeds.2022-006755 [DOI] [PubMed] [Google Scholar]
- 4.Makelarski JA, Thorngren D, Lindau ST. Feed first, ask questions later: alleviating and understanding caregiver food insecurity in an urban children’s hospital. Am J Public Health. 2015;105(8):e98-e104. doi: 10.2105/AJPH.2015.302719 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lee AM, Lopez MA, Haq H, et al. Inpatient food insecurity in caregivers of hospitalized pediatric patients: a mixed methods study. Acad Pediatr. 2021;21(8):1404-1413. doi: 10.1016/j.acap.2021.04.018 [DOI] [PubMed] [Google Scholar]
- 6.van den Heuvel M, Fuller A, Zaffar N, Li X, Beck CE, Birken CS. Food insecurity during COVID-19 in a Canadian academic pediatric hospital: a cross-sectional survey. CMAJ Open. 2022;10(1):e82-e89. doi: 10.9778/cmajo.20210223 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
