Supervising sexual offenders in the community

Research summary
Vol. 12 No. 5
September 2007


How can community supervision officers tell whether sexual offenders are at risk for reoffending?


Once convicted, sexual offenders in Canada are typically managed with a combination of incarceration, treatment and community supervision. For supervision to be effective, however, it is necessary that the officers monitor and address factors related to recidivism.

Although static risk factors are useful for estimating long term recidivism potential, community supervision officers need to monitor the ongoing changes in the offenders' problems that increase their chances of reoffending. These dynamic (changeable) risk factors can be further divided into stable risk factors and acute risk factors. The stable risk factors are relatively enduring traits, such as intimacy deficits; acute risk factors are those that change rapidly, such as intense anger.

Recent research has identified a number of promising dynamic risk factors for sexual offenders. There was a need, however, for a method of assessing and monitoring these factors during community supervision.


The Dynamic Supervision Project began by constructing a method of assessing static, stable and acute risk factors, which was then taught to community supervision officers. Subsequently, officers used these assessment procedures during the routine supervision of the sexual offenders. The static factors were assessed at the beginning of supervision, the stable factors were assessed every six months, and the acute factors were assessed with every supervision contact (about once a month).

The 997 offenders and 156 officers participating in this project came from 16 different jurisdictions, including all Canadian provinces and territories, the Correctional Service of Canada, and the states of Alaska and Iowa. After an average follow-up period of 40 months, 7% reoffended with a sexual crime, 14% reoffended with a sexual or violent crime, and 28% reoffended with any new offence. These rates are on the low end of the expected range, which is 10% to 15% sexual recidivism after 5 years.


The community supervision officers were able to learn the measures, as demonstrated by acceptable levels of agreement between their ratings and the ratings of experts. The original versions of the stable and acute measures had adequate levels of predictive validity, but they were not optimal.

Consequently, new versions included small changes to the definitions of the items, and different methods of combining the items into overall risk categories. The revised risk categories were strongly related to the risk of recidivism, particularly when used by conscientious officers.

The measures were also perceived as sufficiently useful in case management that they were adopted by diverse jurisdictions in Canada, the US and Europe prior to the completion of the evaluation study.

Policy implications

  1. Structured risk tools should be used by community supervision officers to distinguish between sexual offenders at high risk to reoffend and those who can be safely managed with minimal interventions.
  2. The risk measures developed in the current study showed sufficient accuracy that their use in the routine supervision of sexual offenders is recommended.
  3. Further research is needed to replicate the findings of the current study and explore other approaches to advancing risk assessments with sexual offenders. The measures developed for the current study fill an important gap, but the results also suggest that they can be improved.


For further information

R. Karl Hanson, Ph.D.
Corrections Research
Public Safety Canada
340 Laurier Avenue West
Ottawa, Ontario K1A 0P8
Phone: 613-991-2840
Fax: 613-990-8295

Date modified: