The term “big data” refers to data sets that are extensive and realistic, but also collected quickly and in a variety of ways. Thanks to technological advances, such data sets are also being compiled more and more frequently in psychotherapy. They usually contain information about patients, clinical impressions and diagnoses recorded by treating physicians and therapists, as well as psychometric data on symptom severity and other important variables that can be collected before, during, and after treatment.

The unusually large amounts of data available for this field of research enable robust statistical analyses that can answer a wide variety of questions. On the one hand, basic clinical psychological research can gain information about the structure, course, and diversity of mental health problems beyond diagnostic categories and improve the psychometric assessment and modeling of mental health impairments.

Practice-oriented research can also benefit from big data. This is achieved by improving known methodological approaches from this field of research through enhanced algorithms and technology-supported data collection. Innovations in the field of machine learning enable the creation of accurate predictions. This allows prognostic information to be derived from patient characteristics which, when processed appropriately, can be used in everyday clinical practice. Probabilistic statements about the expected course of therapy are just as conceivable as an assessment of the risk of discontinuation. Regularly collected data on the course of therapy can be evaluated and reported back in real time. Methods for personalizing psychotherapeutic interventions, which were previously based on secondary analyses of smaller data sets, are becoming more robust and ecologically valid.

Cooperation partners: