OBJECTIVE: To determine the association between self-reported diabetes and posttraumatic stress disorder (PTSD) among adults in the community and to investigate the specificity of this link. METHOD: Data were drawn from the National Comorbidity Survey (NCS), a household probability sample of adults ages 15-54 in the United States. Multiple logistic regression analyses were used to investigate the association between self-reported diabetes and PTSD (past 12-month prevalence), adjusted for differences in sociodemographic characteristics. RESULTS: Self-reported diabetes was associated with an increased likelihood of PTSD [OR = 2.3 (1.02, 5.21)], which persisted after adjusting for differences in sociodemographic characteristics. Self-reported diabetes was not associated with a significantly increased likelihood of any other mental disorders. CONCLUSIONS: These findings are consistent with and extend previous data suggesting that there is an association between self-reported diabetes and PTSD by showing that this link appears to have some specificity and is generalizable to adults in the community. These data do not address the possible mechanisms of this association. Our results do not support a link between depression and diabetes, which has been previously reported. Replication of these results is needed with longitudinal, epidemiologic data, which include ages of onset and physiologic data in diagnosis of diabetes. If these findings are replicated, further investigation into the possible mechanisms of this association may be fruitful.