Objective: This prospective study assessed the use of seclusion and restraint in the Pennsylvania state hospital system from 2001 through 2010. It also examined the correlation between declining use of containment procedures and assaults by patients on other patients and staff.Methods: The 12,900 anonymized records involving the 1,801 unique, civilly committed individuals who were physically or mechanically restrained and secluded in the nine civil hospitals during this study period were entered into a database. The data set included demographic and diagnostic information about the patients and the cause and effect of the procedures. These data were compared with rates of patient-to-patient and patientto-staff assaults to determine any correlation between changes in use of containment and assaults.Results: From 2001 to 2010, the use of mechanical restraint significantly declined from .37 to .08 episodes per 1,000 days (p,.018), and the use of seclusion significantly declined from .21 to .01 episodes per 1,000 days (p,.001). Persons with an axis I diagnosis of psychotic disorder accounted for 44% of containment procedures used during this study. Patient-topatient assaults declined slightly, and patient-to-staff assaults were unaffected.Conclusions: Decreasing the use of containment procedures did not increase assaults. Better leadership, data transparency, use of clinical alerts, workforce development, policy changes, enhanced use of response teams, implementation of dialectical behavior therapy, and discontinuation of the psychiatric use of PRN orders contributed to the change in use of containment procedures. A philosophical change to a recovery model of psychiatric care and services was the driving force behind this transformation.