OBJECTIVE: To test the effectiveness of an integrative psychotherapeutic nursing home program (integrative reactivation and rehabilitation [IRR]) to reduce multiple neuropsychiatry symptoms (MNPS) of cognitively impaired patients and caregiver burden (CB). DESIGN: Randomized controlled trial. SETTING: Psychiatric-skilled nursing home (IRR) and usual care (UC), consisting of different types of nursing home care at home or in an institution. PARTICIPANTS: N = 168 (81 IRR and 87 UC). Patients had to meet classification of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition for dementia, amnestic disorders, or other cognitive disorders. Further inclusion criteria: Neuropsychiatric Inventory (NPI) ≥3; Mini-Mental State Examination ≥18 and ≤27; and Barthel Index (BI) ≥5 and ≤19. INTERVENTION: IRR consisted of a person-oriented integrative psychotherapeutic nursing home program to reduce MNPS of the patient and CB. UC consisted of different types of nursing home care at home or in an institution, mostly emotion oriented. MEASUREMENTS: Primary outcome variable was MNPS (number and sum-severity of NPI). Furthermore, burden and competence of caregiver were also measured. ASSESSMENTS: T1 (inclusion), T2 (end of treatment), T3 (after 6 months of follow-up). Cohen's d (Cd) was calculated for mean differences (intention to treat). For confounding, repeated measurement modeling (random regression modeling [RRM]) was applied. RESULTS: In the short term from the perspective of the caregiver, IRR showed up to 34% surplus effects on MNPS of the patients; NPI symptoms: 1.31 lower (Cd,-0.53); and NPI sum-severity: 11.16 lower (Cd,-0.53). In follow-up, the effects were sustained. However, from the perspective of the nursing team, these effects were insignificant, although the trend was in the same direction and correlated significantly with the caregiver results over time (at T3: r = 0.48). In addition, IRR showed surplus effects (up to 36%) on burden and competence of caregiver: NPI emotional distress: 3.78 (Cd,-0.44); CB: 17.69 (Cd,-0.63) lower; and Competence: 6.26 (Cd, 0.61) higher. In follow-up, the effects increased up to 50%. RRM demonstrated that the effects were stable. CONCLUSION: From the perspective of the caregiver, IRR was significantly more effective than UC to reduce MNPS in cognitively impaired patients and CB. In follow-up, the effect on CB even increased. However, from the perspective of the nursing team, the effects on MNPS were statistically insignificant. Nevertheless, the trend was in the same direction and correlated significantly with the caregiver results over time. Further research is needed, preferably using a blinded randomized controlled trial. © 2011 American Association for Geriatric Psychiatry.