Assessing ketamine-related risk: development and inter-rater reliability of the KASPR tool

Abstract

Background Ketamine use has increased in recent years, with growing recognition of associated physical, psychological, and cognitive harms. Existing assessment approaches, largely derived from dependence-based models, may be insufficiently sensitive to the multi-domain and progressive nature of ketamine-related harm, particularly in its earlier stages.Methods A vignette-based inter-rater reliability study was conducted in which 25 practitioners applied the Ketamine Assessment, Screening and Progression Risk (KASPR) tool to eight standardized case scenarios. KASPR is a clinician-completed, 11-item assessment organized across Exposure, Drivers, and Harm. Domain and total scores were calculated from item-level responses. Inter-rater reliability was assessed using intraclass correlation coefficients (ICC) and Fleiss’ kappa.Results KASPR demonstrated good overall inter-rater reliability for total scores (ICC(2,1) = 0.895). Agreement was high for Exposure (ICC = 0.909) and Harm (ICC = 0.941) and lower for Drivers (ICC = 0.585). Agreement on intervention tier (kappa = 0.324) and primary clinical focus (kappa = 0.388) was fair.Conclusions KASPR demonstrated good inter-rater reliability in a multidisciplinary practice-based sample. A domain-based approach incorporating sentinel symptoms may support more consistent recognition of ketamine-related risk, although lower agreement for psychologically interpretive and categorical decisions indicates a need for further evaluation in real-world clinical settings.