Background: Alcohol-related disorders (ARD) encountered in emergency departments (ED) have a high prevalence and are underestimated. It is necessary to provide professionals with a tool to identify patients in whom there is a risk that alcohol-related and mental health problems may be associated. Kessler's K6/10 psychological distress scales are fast, easy-to-use, and have been shown to achieve a good performance in the identification of psychological distress associated with ARD.
Aim: The aim of this study was to evaluate the psychometric properties of the Kessler scales, version 6 and 10, with a sample of patients admitted to EDs for alcohol consumption.
Methods: On the day after their admission, with a zero "blood" alcohol concentration, 71 patients were randomly assigned to be assessed using 6 or 10 items version. The internal consistency and factor structure of the K6/10 versions were examined. Convergent validity was measured using the Hospital Anxiety and Depression Scale (HADS) and the Hamilton Depression Rating Scale (HDRS).
Results: The prevalence of psychological distress in our sample was approximately 60%. The selected threshold scores were 10 for K6 (Sensitivity: 0.92; Specificity: 0.62) and 14 for K10 (Sensitivity: 0.95; Specificity: 0.54). The Cronbach coefficients for K6 and K10 were 0.76 and 0.84, respectively. The factor analyses indicated the multidimensional nature of K6/10. The 2 versions, containing 6 and 10 items respectively, correlated better with the HADS (0.83 and 0.70, respectively) than with the HDRS (0.51 and 0.49, respectively). The areas under the ROC Curve indicated a high level of accuracy for both the K6 (0.87) and the K10 (0.77). The difference was not statistically significant.
Conclusions: This study confirms the good psychometric characteristics of Kessler's psychological distress scale. Even though similar performances were observed for K6/10, the brevity of the K6 makes it more suitable for use in EDs.