Persons suffering from mental illness, retardation or other disturbances who
commit crimes in Denmark cannot be punished, but are instead sentenced to
treatment. In the past, these sentences were always indeterminate, meaning that
treatment regimens were only terminated when psychiatrists judged these persons no longer a danger to themselves or others. This practice was, however, severely criticized by advocates for the mentally ill in the late 1990s, since many of these treatment sentences lasted far longer than penal sentences given to mentally competent persons found guilty of similar crimes. This critique, which hinges on proportionality, led to a change in the law on July 1, 2000, after which the length of treatment for non-serious-violent crimes was limited to a maximum of five years. This change was itself criticized by the Council of Medical Forensic Specialists, who argued that time of release should be based on mental health criteria as opposed to simple time schedules, and that release after five years would not provide adequate time to treat all persons – implying the possibility of a heightened risk of recidivism. The current study compares rates of recidivism among mentally disturbed persons sentenced before and after the July 1, 2000 change in the law. The analysis is based on 1510 persons serving 1554 sentences and utilizes data from the criminal register, the central psychiatric register, the cause of death register, and the population register. The data do not support the contention that a determinate treatment sentencing regime increases the risk of recidivism.