Background: Psychosis is a common phenomenon in Alzheimer's disease (AD). The APOE
ε4 allele is the strongest genetic risk factor for the development of AD, but its association with psychosis
Objective: We investigated the associations between psychosis, subdivided into delusions and hallucinations,
as well as APOE ε4 allele on cognitive and functional outcomes. Secondarily, we investigated the
associations between APOE ε4, Lewy bodies, and psychosis.
Methods: Data from the National Alzheimer's Coordinating Center (NACC) were used. Nine hundred
patients with a confirmed diagnosis of AD based on the NIA-AA Reagan were included in the analysis.
Global cognition was assessed using the Mini-Mental State Exam (MMSE) and functional status was
assessed using the Functional Activities Questionnaire (FAQ). Psychosis status was determined using
the Neuropsychiatric Inventory Questionnaire (NPI-Q). Factorial design was used to assess the effects of
psychosis and APOE ε4, as well as their interaction.
Results: Psychosis and the presence of APOE ε4 were both associated with lower MMSE scores, while
only psychosis was associated with higher FAQ scores. Furthermore, patients with hallucinations had
lower MMSE and higher FAQ scores than patients with only delusions. There was a significant interaction
effect between psychosis and APOE ε4 on MMSE scores, with APOE ε4 negatively affecting patients
with hallucinations-only psychosis. APOE ε4 was positively associated with the presence of Lewy
body pathology, and both were found to be more prevalent in psychotic patients, with a stronger association
Conclusion: Psychosis in AD was associated with greater cognitive and functional impairments. Patients
with hallucinations-with or without delusions-conferred even greater deficits compared to patients
with only delusions. The APOE ε4 allele was associated with worse cognition, especially for patients
with hallucination-only psychosis. APOE ε4 may mediate cognitive impairment in the hallucinations
phenotype through the development of Lewy bodies. Our findings support that subtypes of psychosis
should be evaluated separately.