SummaryIn this project, we propose to apply a previously established and methodologically validated cluster analysis approach for assessing constellations of child maltreatment (CM), to the large dataset of Childhood Trauma Screener (CTS) information in the German National Cohort. This novel analysis uses an unsupervised clustering approach to integrate information on co-occuring CM dimensions (emotional neglect, physical neglect, emotional abuse, physical abuse, sexual abuse) and has shown promising results in the interpretation of Childhood Trauma Questionnaire (CTQ) data. This approach has hitherto not been applied on the much shorter CTS. We aim to validate these clusters of co-occurring CM and evaluate those as predictors for psychopathological trajectories as well as vulnerability factors for a comprehensive array of somatic conditions. The clusters will also be charaterised with a variety of biological markers and used in a supervised machine learning algorithm to predict disease onset within 2,5 years after baseline assessment.

Keywords-

InstitutionsLMU Klinikum, Helmholtz Zentrum München, Klinikum der Universität München, Universitätsmedizin Greifswald, LMU Klinikum München, Klinik für Psychiatrie und Psychotherapie