The Effectiveness of Inpatient Geriatric Evaluation and Management Units: A Systematic Review and Meta-Analysis

OBJECTIVES To examine how geriatric evaluation and management units (GEMUs) are organized and to examine the effectiveness of admission on a GEMU. DESIGN Systematic review and meta-analysis based on literature search of multiple databases and the references lists of all identified articles and by contacting authors. SETTING GEMUs. PARTICIPANTS Elderly people admitted to a GEMU. MEASUREMENTS Quality of the studies was assessed on 10 criteria. The outcome parameters were mortality, institutionalization, functional decline, readmission, and length of stay at different follow-up points. A random-effects meta-analysis was performed using Hedges' gu and variance of relative risk (RR). RESULTS GEMUs are organized in a heterogeneous way and the included studies gave no thorough description of comprehensive geriatric assessment (CGA). Involvement of a multidisciplinary team was a key element in all GEMUs. The individual trials showed that admission to a GEMU has one or more favorable effects on the outcomes of interest, with two significant effects in the meta-analysis: less functional decline at discharge from the GEMU (RR=0.87, 95% confidence interval (CI)=0.77-0.99; P=.04) and a lower rate of institutionalization 1 year after discharge (RR=0.78, CI=0.66-0.92; P=.003). For the other outcomes in the meta-analysis, a GEMU did not induce significantly different outcomes than usual care. CONCLUSION This meta-analysis shows a significant effect in favor of the GEMU group on functional decline at discharge and on institutionalization after 1 year. There is heterogeneity between the studies, poor quality of some randomized controlled trials, and shortage of information about CGA. Multidisciplinary CGA offered in a GEMU may add value to the care for frail older persons admitted to the hospital, but the limitations confirm the need for well-designed studies using explicit CGA and more-structured and -coherent assessment instruments such as the Minimum Data Set Resident Assessment Instrument.