Abstract Introduction: There has been a significant increase in the use of testosterone in aging men, but little investigation into its impact on men with Alzheimer's disease (AD). The findings of the few studies that have been done are inconsistent. In the present study, we investigated the relationship between total testosterone (TT) and neuropsychiatric symptoms (NPS) in a well-characterized sample of elderly men with mild to moderate AD. Methods: The sample, which was drawn from the Texas Alzheimer's Research Care Consortium Longitudinal Research Cohort, included 87 men who met the criteria for mild to moderate AD. The occurrence of NPS was gathered from caregivers and/or family members with the Neuropsychiatric Inventory. TT was analyzed, and the sample was divided into a low-testosterone group (TT ≤2.5 ng/ml; n = 44) and a borderline/normal group (TT ≥2.6 ng/ml; n = 43). Results: TT was correlated with symptoms of hallucinations, delusions, agitation, irritability and motor activity. The borderline/normal group was significantly more likely to have hallucinations (odds ratio (OR) = 5.56), delusions (OR = 3.87), motor activity (OR = 3.13) and irritability (OR = 2.77) than the low-testosterone group. Health status and apolipoprotein E ε4 status were not significant factors. Conclusions: The findings of the present study have implications for the use of testosterone replacement therapy in men with AD or the prodromal stage of the disease.