OBJECTIVES To determine whether a multidisciplinary team intervention minimizes unanticipated transitions from assisted living for persons with dementia. DESIGN Randomized trial. SETTING Two dementia-specific assisted living facilities in Connecticut owned and managed by the same corporation. PARTICIPANTS One hundred older adults with dementia who relocated to assisted living. INTERVENTION Four systematic multidisciplinary assessments by a geriatrician, geriatrics advanced practice nurse, physical therapist, dietitian, and social worker during the first 9 months of relocation to assisted living. MEASUREMENTS Permanent relocation to a nursing facility, emergency department (ED) visits, hospitalization, and death. RESULTS Fifty-five residents experienced any unanticipated transition out of assisted living, on average 84 +/- 74 days after relocation; falls were the primary reason for transition. The intervention reduced the risk of any unanticipated transitions (13%), permanent relocation to a nursing facility (11%), ED visits (12%), hospitalization (45%), and death (63%), but the results did not meet statistical significance. In secondary analysis, more men experienced any unanticipated transition (P<.001), hospitalization (P<.001), or death (P<.001) than women. CONCLUSION Although an untargeted multidisciplinary intervention did not significantly reduce the risk of transitions for individuals with dementia relocating to assisted living in this small sample, trends for decreasing hospitalization and death were found. The data further suggest that those at risk for falls and men may benefit from targeted clinical interventions to prevent unanticipated transitions, especially during the first 3 months after relocation.