The main objective of the present study was to investigate the impact of treatment on forensic psychiatric inpatients, examining changes on 22 indicators of five dynamic risk factors for violence (i.e., egocentrism, hostility, impulsivity, lack of insight, and negative distrustful attitudes), and to relate these potential changes to level of psychopathy assessed with the Hare Psychopathy Checklist - Revised (PCL-R). Also, we studied the relationship between psychopathy and treatment compliance, as indicated by the attendance rate of therapeutic activities. Eighty-seven male patients (due to missing data on at least one measure, sample size varies from 58 to 87; 42 patients have complete datasets) were administered a standardized psychological assessment battery (self-report inventories, performance-based personality test, observer ratings) upon admission (T1) and after on average 20 months of treatment (T2). Upon admission, psychopathy (median split, PCL-R score >= 22) was significantly related to a higher score on five of the 22 indicators of dynamic risk The analyses showed no significant differences between psychopathic and non-psychopathic patients on the indicators of dynamic risk factors during 20 months of inpatient forensic psychiatric treatment. However, psychopaths showed the expected pattern of treatment noncompliance, compared to non-psychopaths. The clinical and research implications of these findings are discussed.