The effectiveness and variation of acute medical units: a systematic review.

Purpose: To evaluate the evidence for the effectiveness of acute medical units (AMUs) compared with other models of care and compare the components of AMU models.

Data Sources: Six electronic databases and grey literature sources searched between 1990 and 2014.

Study Selection: Studies reporting on AMUs as an intervention for unplanned medical presentations to hospital with the inclusion of all outcome measures/study designs/comparators.

Data Extraction: Data on study characteristics/outcomes/AMU components were extracted by one author and confirmed by a second.

Data Synthesis: Seventeen studies of 12 AMUs across five countries were included. The AMU model was associated with a reduction in-hospital length of stay (LOS) in all analyses ranging from 0.3 to 2.6 days; and a reduction in mortality in 12 of the 14 analyses with the change ranging from a 0.1% increase to a 8.8% reduction. Evidence relating to readmissions and patient/staff satisfaction was less conclusive. There was variation in the following components of AMUs: admission criteria, entry sources, functions and consultant work patterns.

Conclusion: This review provides evidence that AMUs are associated with reductions in-hospital LOS and, less convincingly, mortality compared with other models of care when implemented in European and Australasian settings. Reported estimates may be affected by residual confounding. This review reports heterogeneity in components of the AMU model. Further work to identify what constitutes the key components of an AMU is needed to improve the quality and effectiveness of acute medical care. This is of particular importance given the escalating demand on acute services.