P01-385 - A short screening for depression (DESC) using Rasch analysis: development and validation of the new instrument

Abstract

Although various instruments for the assessment of depression are available, substantial psychometric problems like multidimensionality or dependence on sample characteristics have been demonstrated. Here, modern test theories, e.g. Rasch Analysis, offer new directions of assessments. So far, Rasch analysis has been applied to the evaluation of diagnostic tools for depression, though neither a concise rating instrument is available consequently using Rasch methodology nor an integration with Structural Equation Modelling (SEM). Therefore, two parallel 10 item depression questionnaires of Depression Screening Version (DESC) were developed by combining Rasch analysis and SEM. Both scales were based on previously established Rasch homogeneous item bank for depression (N=367) and proved one-dimensionality according to Rasch and SEM analyses, justifying total score interpretation of the DESC. The new data of 333 patients suffering from psychic disorder including depression as well as different somatic disorder were included for a first psychometric evaluation. Both DESC versions feature different but highly correlated item sets. Cut-off scores with good sensitivity (0.82, 0.81) and specificity (0.78, 0.77) for the diagnosis of an depressive syndrome were developed with ROC analyses. Sensitivity was even superior to the Beck Depression Inventory (BDI). No serious Differential Item Functioning in Rasch analyses for both scales was found. The DESC represents a short screening instrument for depressive symptoms in two parallel versions with highly sufficient psychometric quality. Particularly clinicians, who aim at screening diverse samples of patients within a limited time frame or repeatedly during the course of their diseases, will benefit from this new screening instrument.