Association between hospital admission either directly or via the emergency department, and readmission rates at 30 days in older adults in two rural hospitals: a retrospective cohort study.

INTRODUCTION: Older patients are frequently re-admitted to the hospital after attending the emergency department (ED). We investigated whether direct admission to the hospital was associated with a lower risk of readmission at 30 days compared to admission via the ED, in patients aged ≥ 75 years. METHODS: Retrospective multicenter cohort study from 01/01/2018 to 31/12/2019, including patients aged ≥ 75 years from two hospitals. Patients admitted directly were matched 1:1 with patients admitted via the ED for center, age category, sex, major diagnosis category, type of stay (medical/surgical), and severity. We compared readmission at 30 days (primary outcome) and length of stay (secondary outcome) between groups. RESULTS: A total of 1486 matched patients with an available outcome measure were included for analysis. We observed no significant difference in 30-day readmission rate between those admitted directly (102/778, 13.1%) and those admitted via the ED (87/708, 12.3%, p = 0.63). There was a significant difference in length of stay between both groups: median 5 days [Q1-Q3: 2-8] vs 6 days [2-11] for direct and ED admissions, respectively (effect size: 0.11, p < 0.001). By multivariate analysis, only moderate to severe denutrition was associated with the risk of readmission at 30 days (Odds Ratio 2.133, 95% Confidence Interval 1.309-3.475). CONCLUSION: The mode of entry to the hospital of patients aged 75 years and older was not associated with the risk of readmission at 30 days. However, those admitted directly had a significantly shorter length of stay than those admitted via the ED.