Aim: To explore if patients admitted to forensic psychiatric care decreased their assessed risk for violence over time, to identify patients who decreased their assessed risk for violence exceptionally well (30% or more) on the Clinical (C) and Risk management (R) scales in the (HCR-20), and to compare them in terms of demographic data.

Methods: The HCR-20 risk assessment instrument was used to assess the risk for violence in 267 patients admitted to a Swedish forensic psychiatric clinic between 1997 and 2010. Their assessments at admission were compared with a second, and most recent, risk assessment.

Results: The risk for violence decreased over time. Demographic criteria had no impact on differences on decreased risk. Only two factors, namely gender and psychopathy showed a difference. Risk factors associated with stress and lack of personal support were the items that turned out to be the most difficult to reduce.

Conclusion: The results show that risk prevention in forensic care does work and it is important to continue to work with risk management. The study highlights the importance of a careful analysis of the patient's risk for violence in order to work with the patient's specific risk factors to reduce the risk.