The relationship between geographical residence and hospitalization of older adults in the United States

Le résumé fourni par la source

Abstract Introduction Older adults (65+ years) in rural areas are at a higher risk of frailty and costly healthcare utilization due to ecological factors, such as inconsistent support for accessing healthcare. We aim to assess the relationship between frailty trajectories and inpatient hospitalizations in rural versus urban communities. Methods We included older adults from the National Health and Aging Trend Study (2011-2020) with complete data for Fried’s frailty phenotype scoring (n = 5,540). The Office of Management and Budget’s criteria defined rurality. Previously, latent growth class analysis defined frailty trajectories (improving, stable, mildly and drastically worsening). Using linked Medicare fee-for-service data, negative binomial regression and marginalized two-part model were used to determine how geographic residence modifies the relationship of inpatient hospitalizations and frailty trajectories using incident rate ratios (IRR). Results The mean age of the sample was 75.3±6.74(SD) years, 60.9% (n = 3,373) female, 18.3% (n = 1,013) living in rural communities, and mean yearly hospitalization cost $1,806 (SD $5,522). Trajectory membership breakdown: 8.6% improving, 42.6% stable, 38.4% mildly and 10.4% drastically worsening. Among those living in rural communities, we observed higher IRR associated with the mild [1.81 95%CI (1.55, 2.11)] and drastic worsening [1.81 (1.21, 2.72)] trajectories. For urban communities, this was only observed for the drastically worsening trajectory [2.05 (1.44, 2.93)]. The relationship between hospitalization rate and frailty trajectories was significantly modified by urban residence (p-value 0.018). Conclusion Urban residence may modify the relationship between hospitalization rate and worsening frailty trajectories. Future analyses examining predictors of utilization may serve as intervention targets for frailty prevention.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.