Abstract
BACKGROUND: Delivering hospital-level care with comprehensive geriatric assessment (CGA) in the home is one approach to deal with the increased demand for bed-based hospital care, but clinical effectiveness is uncertain. OBJECTIVE: To assess the clinical effectiveness of admission avoidance hospital at home (HAH) with CGA for older persons. DESIGN: Multisite randomized trial. (ISRCTN registry number: ISRCTN60477865). SETTING: 9 hospital and community sites in the United Kingdom. PATIENTS: 1055 older persons who were medically unwell, were physiologically stable, and were referred for a hospital admission. INTERVENTION: Admission avoidance HAH with CGA versus hospital admission with CGA when available using 2:1 randomization. MEASUREMENTS: The primary outcome of living at home was measured at 6 months. Secondary outcomes were new admission to long-term residential care, death, health status, delirium, and patient satisfaction. RESULTS: Participants had a mean age of 83.3 years (SD, 7.0). At 6-month follow-up, 528 of 672 (78.6%) participants in the CGA HAH group versus 247 of 328 (75.3%) participants in the hospital group were living at home (relative risk [RR], 1.05 [95% CI, 0.95 to 1.15]; P = 0.36); 114 of 673 (16.9%) versus 58 of 328 (17.7%) had died (RR, 0.98 [CI, 0.65 to 1.47]; P = 0.92); and 37 of 646 (5.7%) versus 27 of 311 (8.7%) were in long-term residential care (RR, 0.58 [CI, 0.45 to 0.76]; P < 0.001). LIMITATION: The findings are most applicable to older persons referred from a hospital short-stay acute medical assessment unit; episodes of delirium may have been undetected. CONCLUSION: Admission avoidance HAH with CGA led to similar outcomes as hospital admission in the proportion of older persons living at home as well as a decrease in admissions to long-term residential care at 6 months. This type of service can provide an alternative to hospitalization for selected older persons. PRIMARY FUNDING SOURCE: The National Institute for Health Research Health Services and Delivery Research Programme (12/209/66).