Alzheimer’s disease with psychosis was defined by the presence of delusions or hallucinations as assessed by the Neuropsychiatric Inventory Questionnaire, a caregiver-based rating scale (4). The Alzheimer’s disease without psychosis group comprised individuals for whom Neuropsychiatric Inventory Questionnaire data were available through the entire 36 months who had no delusions or hallucinations. Sixty case subjects who had CSF collected at baseline developed psychosis over the 36-month study period; 13 were psychotic at the first visit, and the remainder developed psychosis over the 36-month period. One hundred fifteen participants had CSF collected at baseline and remained nonpsychotic over the course of the study. Analyses of covariance on each of the three CSF measures were conducted comparing psychotic and nonpsychotic individuals. Age, gender, education, diagnosis at baseline, and global score on the Clinical Dementia Rating Scale were entered as possible covariates; if a covariate did not reach significance, it was deleted. Interactions with psychosis were entered into the analyses; if they were not significant they were deleted from the model. If psychosis was not significant, an analysis of variance was conducted, with the covariate becoming the main independent variable.