Purpose: Many older adults living in nursing homes experience varying degrees of physical and cognitive impairment. It is also estimated that 30%-84% of residents in nursing homes exhibit some form of behavior problem. Whereas there is evidence that behavioral symptoms are associated with physical and cognitive impairments, the underlying mechanisms of the relationship are not clear. The purpose of this study was to longitudinally explore the relationship between cognitive ability, ADLs, and behavioral symptoms of residents in nursing homes over a year period.

Methods: Data for this study were derived from the NINR/NIH funded project, which collected information from residents in five nursing homes in the Pittsburgh area. The sample consists of 78 residents whose behavioral symptoms were measured 3 times (6 months apart up to 12 months). Behavioral symptoms were measured through both staff reporting and computer-assisted direct observation, and cognition and ADLs were reported by staff. Recursive path models were used to examine patterns of the relationship using Amos 5.0.1.

Results: Both models (with staff reporting or direct observation of behavioral symptoms) showed a decent fit. In the model with staff-reported behavioral symptoms, cognition, ADLs, and behavioral symptoms were highly correlated at baseline and predicted worsening of functional and behavioral symptoms (2=25.5; df=22; p=0.28; CFI=0.99; NFI=0.94; RMSEA=0.045). The model with the direct observation demonstrated that cognition and ADLs were more related with each other than with behavioral symptoms and predicted each other. Behavioral symptoms were predicted most by their own lagged values (2=39.7; df=24; p=0.023; CFI=0.99; NFI=0.93; RMSEA=0.09).

Implications for Research and Practice: Recently, non-pharmacologic interventions have gained much attention in treating older adults with dementia. Researchers have demonstrated that those interventions can delay decline of cognition, ADLs, and behavioral symptoms. Findings of this study corroborate that dimensions of intervention in older adults with dementia should include cognition, ADLs, and behavioral symptoms. The results also support the notion that behavioral symptoms tend to be stabilized or to be perceived decreasing with the dementia progress. In such a case, behavioral symptoms alone may not be a good indicator of dementia progress and associated functional and cognitive decline. Close monitoring of change in ADLs and cognition along with behavioral symptoms would facilitate appropriate treatment and intervention for older adults with dementia.