The purpose of this study is to examine the efficacy of non-pharmacological ways to treat agitated behaviors in nursing home residents with dementia. The hypothesis is that non-pharmacological interventions will be more effective than placebo in decreasing such behaviors.

Clinicians often refer to elderly persons under their care as "agitated." However, agitation is not a diagnosis, but rather, a descriptive term pertaining to a group of behavioral signs and symptoms, defined as socially inappropriate verbal, vocal, or motor (physical) activity that is not judged by an outside observer to result directly from the needs or confusion of the agitated individual. While the literature reports positive effects of non-pharmacological interventions, most studies tend to be based on small samples and do not yield statistically significant results. This study examines the impact of non-pharmacologic interventions tailored through an algorithm labeled TREA (Treatment Routes for Exploring Agitation) using a large, well-controlled, systematic research design.

TREA is an objective, systematic method for developing individualized non-pharmacological treatment plans based on an analysis of the agitated person's unmet needs, past and current preferences, past role-identity, cognitive, mobility, and sensory abilities/limitations, and possible causes for particular agitated behaviors. The methodology calls for ascertaining the type of agitated behavior and the most likely etiology, and then matching the intervention to the etiology and to the participant's characteristics.

The protocol involves the following steps:

• Baseline assessment - using standardized assessments (e.g., MMSE, ABMI, CMAI, Self-identity in dementia), collecting background demographic and medical information as well as systematic observations of the agitated resident
• Intervention exploration - after analyzing potential unmet needs as well as preferences, abilities and identities of each participant, investigators develop a list of potentially successful interventions, examining the effectiveness of a wide range of interventions, from family videos to manipulatives (e.g., puzzles, a tool kit)
• Intervention (treatment) phase - Interventions are systematically provided during the hours of highest agitation (determined at baseline) and observations are conducted at the same time. The treatment phase lasts for 10 days. The placebo control group includes a presentation to staff members as to how to intervene with behavior problems and similar observations of residents.
• Follow-up phase includes repeated assessments without interventions.

Inclusion Criteria:

• Age 60 or older
• Diagnosis of dementia derived from the resident's medical chart at the facility or from an attending physician, as based on DSM-IV criteria and the Report of the NINCDS-ADRDA
• Resident of the facility for at least three weeks so that nursing staff members know the resident well enough to accurately assess him or her
• Identified by nursing staff to exhibit either verbal agitation or physical non-aggressive agitation at least several times a day

Exclusion Criteria:

• Lifelong diagnosis of schizophrenia
• Bipolar disorder diagnosed prior to onset of dementia
• Diagnosis of premorbid mental retardation
• Judged by direct-care nursing staff to have a life expectancy of less than 3 months
• Expected to leave the nursing home (either to enter the hospital or to go home) within the next 4 months
• Agitation manifested less than 6 times a day