ABSTRACT Alzheimer's dementia (AD), the most prevalent neurodegenerative disease of aging, affects cognition, emotion, and behavior. Agitation is a common behavioral syndrome that frequently emerges during middle to late stage AD and is characterized by psychomotor hyperactivity, aggression, irritability, yelling, resistance to care, and insomnia. The untoward consequences of agitation and related behavioral disturbances are considerable and include impaired quality of life, accelerated cognitive decline, heightened risk of institutionalization, and increased morbidity and mortality. Agitation also increases caregiver burden, including stress and deleterious health consequences. However, despite the damaging impact of agitation on the patient and caregiver, current treatments have only modest efficacy. Behavior management strategies are widely employed, but effective only in mild cases. Antipsychotics, the most commonly used class of medication for agitation and psychosis in dementia, have demonstrated mixed results in controlled studies and are associated with elevated morbidity and mortality. Thus, there is a clear need for improved interventions, particularly for severe agitation in AD. Electroconvulsive Therapy (ECT) is a safe and effective intervention for severe mood disorders in later life, including depression complicated by psychosis, mania or catatonia. Concerns regarding adverse cognitive effects of ECT, however, have limited ECT's clinical use in treating dementia with agitation. Both retrospective and prospective studies conducted by our group support the safety and efficacy of ECT in patients with AD and severe agitation. ECT, therefore, may represent an effective treatment of severe agitation in AD. We propose a five-site, randomized, single-blind, controlled clinical trial to determine the safety and efficacy of ECT plus usual care compared with Simulated ECT (S-ECT) plus usual care in 200 hospitalized individuals with moderate to severe stage AD , probable Alzheimer's type (based on NIA-AA criteria), complicated by severe agitation. Subjects will be randomized to either (1) ECT for three weeks (up to 9 ECT treatments) plus usual care (UC), defined as standard behavioral therapy and pharmacotherapy or (2) Simulated ECT (S-ECT) plus UC. Primary efficacy will be measured with the Cohen-Mansfield Agitation Inventory (CMAI). Safety parameters include daily assessment of delirium (Confusion Assessment Method, CAM), cognition (Severe Impairment Battery, SIB-8) and serious adverse events. A 12-week follow-up includes monthly assessments to explore stability of agitation reduction.