Effect of antipsychotic withdrawal on behavior and sleep/wake activity in nursing home residents with dementia: a randomized, placebo-controlled, double-blinded study. The Bergen District Nursing Home Study

OBJECTIVES: To explore the effect on sleep/wake activity and on behavioral and psychological symptoms of the withdrawal of antipsychotic medications from nursing home (NH) patients with dementia. DESIGN: Randomized, placebo-controlled, double-blind trial. SETTING: NHs in Bergen, Norway. PARTICIPANTS: Thirty patients (mean age 83.5) taking haloperidol, risperidone, or olanzapine for nonpsychotic symptoms. Intervention: Study participants were randomly assigned to withdrawal (intervention group) or continued treatment with antipsychotic medications (reference group) for 4 consecutive weeks. MEASUREMENTS: Behavioral rating using the Neuropsychiatric Inventory Questionnaire (NPI-Q) and actigraphy. RESULTS: After antipsychotic withdrawal, behavioral scores remained stable or improved in 11 of 15 patients, whereas four had worsening scores. Actigraphy revealed decreased sleep efficiency after drug discontinuation and increased 24-hour and night activity in both groups. Actigraphy records of nighttime and daytime activity indicated sleep problems and restlessness, in terms of the NPI-Q. One patient was restarted on antipsychotics. CONCLUSION: Antipsychotic drug withdrawal affected activity and sleep efficiency over the short term. Increases in total activity and impaired sleep quality after drug discontinuation should be monitored, because the long-term effect of these changes is not known. The NPI-Q and actigraphy are feasible tools that disclose relevant changes occurring during antipsychotic withdrawal in NH patients with dementia. Their use in clinical practice should be substantiated by larger studies.