Forensic patients are occupying an increasingly large number of beds in state psychiatric hospitals. The presence of these mentally ill offenders has raised concerns about the risk they present to nonforensic patients. This study compared the rate of assaults and factors associated with assaultive behavior among 308 nonforensic patients and two groups of forensic patients including 469 patients found not guilty by reason of insanity and 76 pretrial patients. Consistent with other studies, nonforensic patients had higher rates of assaults than either group of forensic patients. However, being a forensic patient did not affect the odds of assault when controlling for the effects of demographic and clinical variables in a multivariate logistic regression analysis. Factors associated with assaults in each of the three patient groups were identified using multivariate analyses. Implications are presented for treatment of assaultive behavior, mixing of forensic and nonforensic patients within state hospitals, forensic release policies, and future research.