Predictors of long-term mortality from a 23-year database of emergency medical admissions

Ir J Med Sci. 2025 Dec 4. doi: 10.1007/s11845-025-04178-x. Online ahead of print.

ABSTRACT

AIM: There is much focus on the management of the acute emergency medical admission, however, long term outcome data following discharge are sparse.

METHODS: We report all the outcome of emergency admissions to an AMAU (Acute Medical Admission Unit) over 23 years (2002-2024). Data was matched to the Irish National Death Register to capture post-discharge mortality. Predictors of long-term mortality were determined from logistic and Cox regression.

RESULTS: The 2002-2024 cohort consisted of 186,004 episodes in 95,192 patients, with 14,092 in-hospital deaths within 30-days. Post discharge, to April 2025, there were an additional 17,808 deaths - 31,900 in total (33.5%) of the admitted cohort. Predictors of long-term mortality were age OR 2.03 (95%CI 1.92, 2.15), AISS OR 1.44 (95%CI 1.40, 1.48), Charlson Comorbidity Score OR 1.28 (95%CI 1.24, 1.32) and a Neurology MDC primary diagnosis 1.11 (95%CI 1.02, 1.19). A high Comorbidity Score, reduced survival half-life to ~ 50. Performance of a blood culture, during the acute admission, predicted worse long-term outcomes, irrespective of its result.

CONCLUSION: The emergency medical admission is a sentinal event, long-term survival appears better than that projected in the literature .

PMID:41343007 | DOI:10.1007/s11845-025-04178-x