Association Between Discharge Destination and Excess Length of Stay in Heart Failure-Related Admissions: A Retrospective Single-Centre Study.

摘要

RATIONALE: Hospital readmissions and excess inpatient length of stay remain important clinical and organisational concerns in cardiovascular care. Although discharge destination has been linked to postacute utilisation and spending, less is known about whether discharge destination is associated with inpatient days exceeding diagnosis-related expected stay. AIMS AND OBJECTIVES: This study examined whether discharge destination was associated with excess length of stay above diagnosis-related expected stay among heart failure-related admissions and whether those additional inpatient days were associated with lower 30-day readmission. METHOD: We conducted a retrospective observational study of administrative admissions recorded between August 2016 and October 2017 at a large academic urban teaching hospital. The primary outcome was excess length of stay, defined as max (length of stay-CMS geometric mean length of stay, 0). The primary explanatory variable was discharge destination, grouped as home, skilled nursing, other facility or discontinued care. Multivariable regression models estimated adjusted associations with excess length of stay, and a secondary generalised estimating equation model assessed 30-day readmission. RESULTS: The final analytic sample included 1076 admissions from 743 unique patients. Mean excess length of stay was 2.47 days, and 24.3% of admissions were followed by a 30-day readmission. In adjusted models, discharge to skilled nursing was associated with 1.46 additional excess days versus home discharge (95% CI: 0.68-2.23; p < 0.001), while discharge to other facility settings was associated with 1.67 additional excess days (95% CI: 0.29-3.04; p = 0.017). Findings were directionally similar in sensitivity analyses. Excess stay was not associated with lower 30-day readmission, and nonhome discharge categories were not associated with lower readmission odds. CONCLUSION: In this single-site observational sample, nonhome discharge, especially skilled nursing placement, was associated with longer stays above DRG-benchmarked expectations without lower short-term readmission. These findings may reflect greater discharge complexity and local postacute transition constraints, although the study did not directly measure the mechanisms underlying prolonged hospitalisation.