Admission rates in a general practitioner-based versus a hospital specialist based, hospital-at-home model: ACCESS, an open-labelled randomised clinical trial of effectiveness

BACKGROUND Hospital at home (HaH) is an alternative to acute admission for elderly patients. It is unclear if should be cared for a primarily by a hospital intern specialist or by the patient's own general practitioner (GP). The study assessed whether a GP based model was more effective than a hospital specialist based model at reducing number of hospital admissions without affecting the patient's recovery or number of deaths. METHODS Pragmatic, randomised, open-labelled multicentre parallel group trial with two arms in four municipalities, four emergency departments and 150 GPs in Southern Denmark, including + 65 years old patients with an acute medical condition that required acute hospital in-patient care. The patients were randomly assigned to hospital specialist based model or GP model of HaH care. Five physical and cognitive performance tests were performed at inclusion and after 7 days. Primary outcome was number of hospital admissions within 7 days. Secondary outcomes were number of admissions within 14, 21 and 30 days, deaths within 30 and 90 days and changes in performance tests. RESULTS Sixty seven patients were enrolled in the GP model and 64 in the hospital specialist model. 45% in the hospital specialist arm versus 24% in the GP arm were admitted within 7 days (effect size 2.7, 95% CI 1.3-5.8; p = 0.01) and this remained significant within 30 days. No differences were found in death or changes in performance tests from day 0-7 days between the two groups. CONCLUSIONS The GP based HaH model was more effective than the hospital specialist model in avoiding hospital admissions within 7 days among elderly patients with an acute medical condition with no differences in mental or physical recovery rates or deaths between the two models. REGISTRATION No. NCT02422849 Registered 27 March 2015. Retrospectively registered.