Preventing Aggression in Veterans With Dementia

Dementia is known primarily for its effects on memory, however, eighty percent of persons with dementia also have behavioral disturbances. This is often not addressed, leading to increased use of nursing homes, higher incidence of injury (both patient and caregiver) and the use of tranquilizing medications. Pain is one of the strongest predictors of aggression. The prevalence of pain in persons with dementia is known to be about 50%. Untreated pain is associated with significant negative outcomes, including increased health care use, inactivity and isolation. The investigators aim to determine whether outcome differences exist between active intervention and control conditions in relation to the occurrence of aggressive behavior, pain and depression, and its impact on pleasant activities, caregiver burden, quality of caregiver-patient relationship, antipsychotic use, health-service use, injuries to patient and caregiver, and nursing home placement. The active intervention, Preventing Aggression in Veterans with Dementia (PAVeD), is a family caregiver-focused, home-based intervention that uses psychoeducational and behavioral approaches to help reduce the risk of aggressive behavior in persons with dementia. The objective of PAVeD is to improve several areas of patient life that are known causes of aggression: pain and distress (including mood problems), lack of pleasurable activities, and difficulty in patient-caregiver communication that may negatively affect relationship quality and caregivers' recognition of pain.

Inclusion criteria

Patients will be eligible to participate in the study if they meet the following criteria:

• have a documented diagnosis of dementia
• receive primary care from the VA
• reside outside a long-term care facility
• live within 45 minutes of the MEDVAMC
• have mild-to-moderate dementia
• have no history of aggression in the past year
• have no evidence of aggression on the CMAI at baseline (i.e., do not score 2 or higher on both frequency and disruptiveness for any of 13 behaviors listed).
• have a caregiver who is directly involved with the patient:

• at least 8 hours per week
• sees the patient at least twice a week
• and speaks English
• report clinically significant pain (either directly or through the caregiver as a proxy)

Exclusion criteria

Patients will be excluded if they have had history of aggression in the past year

• The investigators will administer the aggression subscale of the Cohen-Mansfield Agitation Inventory (CMAI)
• Aggression will be considered present if any of the following items are endorsed as having occurred over the prior year:

• spitting
• cursing/verbal aggression
• hitting
• kicking
• grabbing
• pushing
• throwing
• biting
• scratching
• hurting self/others
• tearing things/destroying property
• making inappropriate verbal sexual advances
• or making inappropriate physical sexual advances

Participants that scored 2 or higher for both frequency and disruptiveness on any of the 13 behaviors listed on the CMAI at baseline were considered aggressive and were excluded from the study.