Food insecurity (FI) has been associated with hospitalization, although the pathways underlying this relationship are poorly understood, in part due to the potential for a bidirectional relationship. This study aimed to determine associations of FI with concurrent and future hospitalization among older adults; mediation by depression and; whether hospitalization increased risk of FI. Participants came from the 2012 and 2014 waves of the Health and Retirement Study (HRS; n = 13,664). HRS is a prospective cohort representative of U.S. adults over the age of 50. Primary analyses included those who were not hospitalized in 2012 (n = 11,776). Not having enough money to buy necessary food or eating less than desired defined food insecurity. The Composite International Diagnostic Interview Short Form provided depression symptomology. Logistic and linear regression examined concurrent and longitudinal associations of FI in 2012 and 2014 with hospitalization in 2014. Path analysis tested mediation of FI with hospitalization frequency by depression symptomology. Finally, logistic regression examined whether hospitalization in 2012 was longitudinally associated with FI in 2014. FI was not associated with future hospitalization (odds ratio (OR) = 1.1; 95% confidence interval (CI) = 0.9-1.4), however; FI was associated with concurrent hospitalization status (OR = 1.4; 95% CI = 1.1-1.8). Depression symptomology explained 17.4% (95% CI = 2.8-32.0%) the association of FI with concurrent hospitalization frequency. Additionally, hospitalization was associated with becoming food insecure (OR = 1.5; 95% CI = 1.2-2.0). Findings may inform best practices for hospital discharge among older adults.
Keywords: internal medicine; nutrition and aging; preventive geriatrics; psychological gerontology; social gerontology.