Association of Fragmented Readmissions and Electronic Information Sharing With Discharge Destination Among Older Adults

• Type of Material: Article
• Authors: Sara Turbow, Emory University; Mohammed Ali, Emory University; Steven Culler, Emory University; Kimberly Rask, Emory University; Molly Perkins, Emory University; Carolyn Clevenger, Emory University; Elizabeth Vaughan, Emory University
• Language: English
• Date: 2023-05-01
• Publisher: JAMA Network Open.
• Publication Version: Final Published Version
• Copyright Statement: 2023 Turbow SD et al. JAMA Network Open.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1001/jamanetworkopen.2023.13592
• Title of Journal or Parent Work: JAMA network open
• Volume: 6
• Issue: 5
• Start Page: e2313592
• End Page: e2313592
• Grant/Funding Information: This study was supported by grant K23 AG065505 from the National Institute on Aging of the National Institutes of Health; the Program for Retaining, Supporting, and Elevating Early-Career Researchers at Emory from the Emory School of Medicine; a gift from the Doris Duke Charitable Foundation; and Georgia clinical translation and science award UL1 TR002378 from the National Institutes of Health.
• Keywords: Cohort Studies; Retrospective Studies; Hospices; Male; Patient Discharge; Female; Medicare; Aged; Patient Readmission; United States; Alzheimer Disease; Humans
• Research Categories: Health Sciences, Nursing

Abstract

Importance: When an older adult is hospitalized, where they are discharged is of utmost importance. Fragmented readmissions, defined as readmissions to a different hospital than a patient was previously discharged from, may increase the risk of a nonhome discharge for older adults. However, this risk may be mitigated via electronic information exchange between the admission and readmission hospitals. Objective: To determine the association of fragmented hospital readmissions and electronic information sharing with discharge destination among Medicare beneficiaries. Design, Setting, and Participants: This cohort study retrospectively examined data from Medicare beneficiaries hospitalized for acute myocardial infarction, congestive heart failure, chronic obstructive pulmonary disease, syncope, urinary tract infection, dehydration, or behavioral issues in 2018 and their 30-day readmission for any reason. The data analysis was completed between November 1, 2021, and October 31, 2022. Exposures: Same hospital vs fragmented readmissions and presence of the same health information exchange (HIE) at the admission and readmission hospitals vs no information shared between the admission and readmission hospitals. Main Outcomes and Measures: The main outcome was discharge destination following the readmission, including home, home with home health, skilled nursing facility (SNF), hospice, leaving against medical advice, or dying. Outcomes were examined for beneficiaries with and without Alzheimer disease using logistic regressions. Results: The cohort included 275 189 admission-readmission pairs, representing 268 768 unique patients (mean [SD] age, 78.9 [9.0] years; 54.1% female and 45.9% male; 12.2% Black, 82.1% White, and 5.7% other race and ethnicity). Of the 31.6% fragmented readmissions in the cohort, 14.3% occurred at hospitals that shared an HIE with the admission hospital. Beneficiaries with same hospital/nonfragmented readmissions tended to be older (mean [SD] age, 78.9 [9.0] vs 77.9 [8.8] for fragmented with same HIE and 78.3 [8.7] years for fragmented without HIE; P < .001). Fragmented readmissions were associated with 10% higher odds of discharge to an SNF (adjusted odds ratio [AOR], 1.10; 95% CI, 1.07-1.12) and 22% lower odds of discharge home with home health (AOR, 0.78; 95% CI, 0.76-0.80) compared with same hospital/nonfragmented readmissions. When the admission and readmission hospital shared an HIE, beneficiaries had 9% to 15% higher odds of discharge home with home health (patients without Alzheimer disease: AOR, 1.09 [95% CI, 1.04-1.16]; patients with Alzheimer disease: AOR, 1.15 [95% CI, 1.01-1.32]) compared with fragmented readmissions where information sharing was not available. Conclusions and Relevance: In this cohort study of Medicare beneficiaries with 30-day readmissions, whether a readmission is fragmented was associated with discharge destination. Among fragmented readmissions, shared HIE across admission and readmission hospitals was associated with higher odds of discharge home with home health. Efforts to study the utility of HIE for care coordination for older adults should be pursued.