Sexual violence risk assessment: an investigation into the inter-rater reliability of the RSVP in Scotland

Sutherland, Alan Allardice (2010) Sexual violence risk assessment: an investigation into the inter-rater reliability of the RSVP in Scotland. D Clin Psy thesis, University of Glasgow.

Abstract

Background: The RSVP (Risk of Sexual Violence Protocol; Hart et al, 2003) is a structured professional judgement tool for assessing risk of sexual violence. Despite being widely used in forensic mental health settings, the reliability and validity of the RSVP has not been adequately established. There is an urgent need to investigate the inter-rater reliability of the tool in a multi-disciplinary clinical context. Method: Clinicians (n=28) with varying professions, levels of experience and training, used the RSVP to evaluate six case vignettes with varying offence characteristics, clinical-complexity and risk. ICC (Intra-class Correlation Coefficient) and percentage agreement statistics were used to evaluate inter-rater reliability of RSVP items, domains and steps. Items included additional forced-choice judgements relating to Scenario Planning and Case Management steps. Clinician responses were also compared to ‘gold-standard’ judgements developed by experts in the field of forensic risk assessment. Results: Inter-rater reliability was ‘fair’ overall with individual items ranging from ‘poor’ to ‘excellent’. Importantly, there was a ‘good’ level of inter-rater reliability on Summary Judgements and Supervision Recommendations. Inter-rater reliability was highest when used by professionals who were highly trained in forensic risk-assessment. On average, professionals with lower levels of specialist training agreed less with their colleagues and experts, and provided higher estimations of sexual violence risk. Lower levels of agreement were found in cases with moderate levels of complexity and risk. Conclusions: The RSVP can be used to attain adequate levels of inter-rater reliability. However this is dependant on the training and expertise of professionals who use the tool. Methodological strengths and limitations are considered, followed by a discussion of implications for training, practice and future research.

• Item Type: Thesis (D Clin Psy)
• Qualification Level: Doctoral
• Keywords: forensic risk assessment, sexual violence, inter-rater reliability, structured professional judgement, sexual offending and Risk of Sexual Violence Protocol.
• Subjects: R Medicine > R Medicine (General); B Philosophy. Psychology. Religion > BF Psychology
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing
• Supervisor's Name: Kate, Prof. Davidson and Lorraine, Dr. Johnstone
• Date of Award: 2010
• Depositing User: Mr Alan Sutherland
• Unique ID: glathesis:2010-2168
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 24 Nov 2010
• Last Modified: 10 Dec 2012 13:52
• URI: https://theses.gla.ac.uk/id/eprint/2168