Development and validation of a short-form (6-item) version of the clinician-administered dissociative states scale (CADSS-SF).

Abstract

Background: State dissociation is commonly assessed using the Clinician Administered Dissociative States Scale (CADSS; 19-items). A briefer CADSS would have many advantages, enabling assessment of transient dissociative states and allowing repeated assessment within short intervals while minimizing participant burden. This is especially relevant in studies of dissociative drugs which cause sedation and psychomotor slowing. Here we describe the process of developing a short-form version of the CADSS (CADSS-SF), based on nitrous oxide (N2O) - induced dissociative responses in healthy volunteers.Methods: In the 'development phase', using data from three experimental pharmacological studies on N2O in healthy volunteers (n = 229), we identified the most 'N2O-responsive' items with the highest item-total correlations, endorsed by experts and consistent with published accounts on the phenomenology of drug-induced dissociation. Identified items were subjected to confirmatory factor analysis using a separate validation dataset (n = 80), which tested a series of one- and two-factor models with 6-8 items.Results: A 6-item, single-factor CADSS-SF consisted of derealization and depersonalization items and showed excellent model fit (χ2(9) = 0.246, p = .246, CFI/TLI>0.99, RMSEA = 0.059). The CADSS-SF was internally consistent (w = 0.87), correlated strongly with the full scale (r ≥ 0.88) and moderately with a measure of the related, but distinct construct of psychotomimesis (r = 0.63).Discussion: The CADSS-SF is a promising tool for rapid assessment of dissociation. It may be useful for capturing fleeting experimentally-induced dissociative phenomena that would otherwise be disrupted through the process of extended self-reporting, or for studying dissociation during drug intoxication, which is often accompanied by psychomotor slowing, sedation and inattention. The scale's brevity may allow tracking of changes in dissociation over relatively brief periods. However, like the parent long-form (19-item) CADSS, the CADSS-SF primarily captures variations in derealization/depersonalization only and may therefore be less appropriate for capturing the multidimensionality of dissociative phenomena. Further validation is required to establish the generalisability of the CADSS-SF beyond experimentally-(drug-) induced dissociation in healthy populations.