30-day hospital admission among older adults managed at home by a mobile emergency unit (MEU): a retrospective cohort study

Background In recent years, Denmark has introduced mobile emergency unit (MEU) to provide patients with home-based evaluation and treatment by emergency medicine physicians. The aim is to avoid unnecessary hospital admissions and to reduce overcrowding in emergency departments. However, it is unknown which demographic, clinical, and paraclinical characteristics of patients at the index MEU assessment are related to subsequent hospital admission. Therefore, we aimed to describe these baseline characteristics and to examine their association with 30-day hospital admission. Methods In this retrospective, single-centre cohort study at Esbjerg Hospital (Region of Southern Denmark), we screened 1,656 MEU contacts (from 1 January to 31 December 2024) and included adults aged ≥ 65 years, who were managed at home. These patients were potential candidates for hospital admission, and the emergency physician made an on-scene decision regarding admission. Data were analysed using multivariable logistic regression. Results We included 357 MEU contacts, with a median (interquartile range) age of 83.5 (11.7) years. 140 (39.2%) of these contacts were admitted to hospital within 30 days. A higher proportion of the admitted patients had a pre-existing do-not-attempt-resuscitation (DNAR) order compared with the non-admitted patients (85.0% vs 66.4%; p  < 0.001) and lived at home (57.8% vs. 47.4%; p  = 0.055). Chronic pulmonary disease was more common among the admitted patients (31.4% vs 19.3%; p  = 0.009), whereas dementia was less frequent (18.6% vs 28.1%; p  = 0.042). Both a pre-existing DNAR order (odds ratio [OR] 3.83, 95% confidence interval [CI] 2.05–7.16) and home (vs nursing home) residence (OR 1.76, 95% CI 1.03–2.98) were significantly associated with hospital admission in the adjusted model. Conclusions Among older adults assessed at home by MEU physicians, a pre-existing DNAR order and home (vs nursing home) residence were independently associated with 30-day hospital admission. These findings may inform triage and follow-up planning. However, prospective studies are required to establish causal links.