UNLABELLED To explore the association between posttraumatic stress disorder (PTSD) and dementia in older veterans. DESIGN Administrative database study of individuals seen within one regional division of the Veterans Affairs healthcare network. SETTING Veterans Integrated Service Network 16. PARTICIPANTS Veterans aged 65 and older who had a diagnosis of PTSD or who were recipients of a Purple Heart (PH) and a comparison group of the same age with no PTSD diagnosis or PH were divided into four groups: those with PTSD and no PH (PTSD+/PH-, n=3,660), those with PH and no PTSD (PTSD-/PH+, n=1,503), those with PTSD and a PH (PTSD+/PH+, n=153), and those without PTSD or a PH (PTSD-/PH-, n=5,165). MEASUREMENTS Incidence and prevalence of dementia after controlling for confounding factors in multivariate logistic regression. RESULTS The PTSD+/PH- group had a significantly higher incidence and prevalence of dementia than the groups without PTSD with or without a PH. The prevalence and incidence of a dementia diagnosis remained two times as high in the PTSD+/PH- group as in the PTSD-/PH+ or PTSD-/PH- group after adjusting for the confounding factors. There were no statistically significant differences between the other groups. CONCLUSION The incidence and prevalence of dementia is greater in veterans with PTSD. It is unclear whether this is due to a common risk factor underlying PTSD and dementia or to PTSD being a risk factor for dementia. Regardless, this study suggests that veterans with PTSD should be screened more closely for dementia. Because PTSD is so common in veterans, this association has important implications for veteran care.