Randomized controlled trial of different models of care for nursing home residents with dementia complicated by depression or psychosis

Objective: To compare the outcomes of 3 interventions for the management of dementia complicated by depression or psychosis: psychogeriatric case management, general practitioners with specialist psycho-geriatric consultation, and standard care for nursing home residents. Method: The sample for this 12-week randomized controlled trial consisted of 86 subjects with dementia from 11 Sydney, Australia, nursing homes, of whom 34 had depression, 33 had depression and psychosis, and 19 had psychosis. All participants received full psychiatric assessments and physical examinations. Information was obtained from the residents' families and nursing home staff. Depression measures included the Even Briefer Assessment Scale for Depression, Hamilton Rating Scale for Depression, Cornell Scale for Depression in Dementia, and Geriatric Depression Scale. Psychosis measures included the Behavioral Pathology in Alzheimer's Disease Rating Scale, Neuropsychiatric Inventory, and Scale for the Assessment of Positive Symptoms. Data were obtained from nursing home records on prescription of psychotropic medication and demographic information. Management plans were formulated by a multidisciplinary team before random assignment to interventions. Results: All 3 groups improved from pretreatment to posttreatment on depression scales for depression groups and psychosis scales for psychosis groups. Mode of management appeared to make no difference in rate or amount of improvement; neither of the treatment group-by-time interactions were significant. Neither use of antidepressants nor use of antipsychotics predicted depression or psychosis outcomes. Conclusion: Participation in the study was associated with improvement in depression and psychosis, perhaps because of the presence of a psychogeriatric team, the increased attention focused on residents, or the generalization of active intervention techniques to control subjects. A formula-driven psychogeriatric team case management approach was not significantly more effective than a consultative approach or standard care. This study demonstrates the difficulties and feasibility of conducting service-oriented research in nursing homes.