Association of Variation in Behavioral Symptoms With Initial Cognitive Phenotype in Adults With Dementia Confirmed by Neuropathology

IMPORTANCE: Behavioral and psychological symptoms of dementia (BPSDs) in association with amnestic and nonamnestic cognitive phenotypes have not been evaluated across diagnoses of Alzheimer disease pathology (ADP), Lewy body–related pathology (LRP), and mixed pathology (ADP-LRP). OBJECTIVES: To determine the clinical phenotypes at the initial visit that are associated with the nature and severity of BPSDs in patients with ADP, LRP, and ADP-LRP. DESIGN, SETTING, AND PARTICIPANTS: This retrospective longitudinal cohort study included 2422 participants with neuropathologically confirmed ADP, LRP, or mixed ADP-LRP in the National Alzheimer Coordinating Center database from June 20, 2005, to September 4, 2019. Participants had a mean (SD) interval of 5.5 (2.8) years from initial visit to autopsy. MAIN OUTCOMES AND MEASURES: Clinician-determined diagnosis of change across 10 BPSDs (agitation, apathy, depression, delusions, disinhibition, auditory hallucinations, visual hallucinations, irritability, personality change, and rapid eye movement [REM] sleep behavior) and the highest severity score for behavioral change on the Neuropsychiatric Inventory Questionnaire (NPI-Q). RESULTS: A total of 2422 participants (1187 with ADP, 904 with ADP-LRP, and 331 with LRP) were included in the analysis (1446 men [59.7%]; mean [SD] age, 74.4 [10.1] years). Compared with initial amnestic symptoms, executive symptoms were associated with a higher risk for 7 of the 10 BPSDs (hazard ratio [HR] range, 1.28-2.45), and visuospatial symptoms were associated with a higher risk for 2 of the 10 BPSDs (HR range, 1.91-2.51), but neither were associated with a low risk for any BPSD. Language symptoms were associated with a low risk of onset for 3 of 10 BPSDs (HR range, 0.43-0.79) and a high risk for 1 BPSD (personality change) (HR, 1.42 [95% CI, 1.10-1.83]). Participants with LRP had a lower risk for agitation (HR, 0.74 [95% CI, 0.60-0.92]), disinhibition (HR, 0.78 [95% CI, 0.62-0.99]), and irritability (HR, 0.81 [95% CI, 0.68-0.96]) and a higher risk for apathy (HR, 1.19 [95% CI, 1.02-1.38]), depression (HR, 1.32 [95% CI, 1.12-1.55]), auditory (HR, 2.00 [95% CI, 1.37-2.93]) and visual (HR, 2.78 [95% CI, 2.21-3.49]) hallucinations, and REM sleep behavior changes (HR, 4.77 [95% CI, 3.61-6.31]) compared with the ADP group. The ADP-LRP group had a higher risk for delusions (HR, 1.27 [95% CI, 1.08-1.48]), auditory (HR, 1.62 [95% CI, 1.21-2.15]) and visual (HR, 1.57 [95% CI, 1.30-1.89]) hallucinations, and REM sleep behavior changes (HR, 2.10 [95% CI, 1.63-2.70]) than the ADP group and a lower risk for visual hallucinations (HR, 0.56 [95% CI, 0.45-0.71]) and REM sleep behavior changes (HR, 0.44 [95% CI, 0.34-0.57) than the LRP group. Overall, women showed a lower risk of agitation (HR, 0.86 [95% CI, 0.75-0.98]), apathy (HR, 0.79 [95% CI, 0.71-0.87]), visual hallucinations (HR, 0.76 [95% CI, 0.64-0.90]), irritability (HR, 0.77 [95% CI, 0.69-0.86]), and REM sleep behavior change (HR, 0.45 [95% CI, 0.35-0.58]) and a higher risk of depression (HR, 1.26 [95% CI, 1.13-1.41]). Older age was associated with a lower risk of most BPSDs (HR range, 0.98-0.99) except delusions (HR, 1.00 [95% CI, 1.00-1.01]) and auditory hallucinations (HR, 0.99 [95% CI, 0.97-1.00]) and a low NPI-Q composite score (β = −0.07 [95% CI, −0.08 to −0.05]; P < .001). CONCLUSIONS AND RELEVANCE: These findings suggest that the risks of BPSDs differ with respect to the initial cognitive phenotype, underlying neuropathology, age, and sex. Awareness of these associations could be helpful in dementia management.