We aimed to examine risk of diabetes mellitus (DM) among older adults with Alzheimer's disease receiving 3 types of psychotropic drugs, that is, antipsychotics, antidepressants, and sedative anxiolytics. We retrospectively analyzed data from a hospital-based Clinical Research Center for Dementia of South Korea (CREDOS) study conducted between January 1, 2008 and December 31, 2012. Participants (n = 3042) with Alzheimer's disease were aged 65 or older and had no preexisting history of DM. Development of DM was identified using claims for initiating at least 1 prescription of antidiabetic medications or a diagnosis of DM during the follow-up period. Cox proportional hazards regression was used to demonstrate the Hazard ratio of DM in use of each psychotropic drug. Among the 3042 participants, 426 patients (14.0%) developed DM, representing an incidence rate of 5.2/100 person-years during an average 2.9 years of follow-up period. Among the 3 types of psychotropic drugs, antipsychotic users had a significantly higher risk of DM (hazard ratio = 1.74, 95% confidence interval = 1.10, 2.76) than nonusers, after adjusting covariates. Antidepressants and sedative anxiolytics did not achieve statistical significance. These results suggested that the diabetes risk was elevated in Alzheimer patients on antipsychotic treatment. Therefore, patients with Alzheimer's disease receiving antipsychotic treatment should be carefully monitored for the development of DM.