There is considerable responsibility on the clinician to identify sex offenders who may potentially commit more serious sexually violent behaviour and an increased demand for evidence based risk assessments (Macpherson, 1997; Thomas-Peter and Warren, 1998). Offenders who commit non-contact sexual crimes are traditionally classified as harmless despite the significant minority who escalate in offence severity towards more violent sexual offending. Forty convicted male sex offenders were classified as non-contact or contact sexual recidivists. Non-contact recidivists had a history of non-contact sexual offending on two or more occasions. Contact recidivists had a history of noncontact offending and had recidivated with a contact sexual offence. Groups were compared on the Sexual Violence Risk-20 (SVR-20: Boer et al. 1997) and the Psychopathy Checklist: Screening Version (PCL: SV: Hart et al. 1995). Psychosexual variables, criminal history and clinical risk factors were also coded using a multi-variable assessment model. A retrospective-prospective comparison successfully used by Quinsey et al. (1995) was performed between non-contact and contact recidivist groups. Factors that discriminated between non-contact recidivists and contact recidivists were primarily historical in nature, reflecting fixed or relatively stable characteristics. Significant differences between non-contact recidivists and contact recidivists were observed on total PCL: SV scores and psychosocial factors of the SVR-20 including sexual deviation, a history of childhood victimisation and past nonviolent offences. Contact recidivists were significantly younger than non-contact recidivists at first non-sexual offence and were significantly more likely to have a history of homosexual offending. A high level of interrater reliability on the SVR-20 and PCL: SV was observed. Suggested revisions to several iten1s of the SVR-20 and methodological considerations are reported. The research demonstrates that a progressive pattern of sexual offending from noncontact sexual offending to contact sexual offending is reliably associated with a combination of risk factors. The study offers the potential for early detection of a more serious escalation in sexual offending to allow for the possibility of supervision and clinical risk management.